Interactive training system for desalination and domestication of young crabs in saline-alkali soil, salinity regulation and control model optimization method, equipment and medium

Abstract

The invention provides a blue crab saline-alkali soil seedling desalination and domestication interactive training system, a salinity regulation and control model optimization method, equipment and a medium, a dynamic physiological response model is obtained by establishing a blue crab seedling salinity stress response model and performing parameter estimation and calibration based on actual breeding data; carrying out global sensitivity analysis according to the dynamic physiological response model, and determining a physiological tolerance interval of key regulatory factors related to the desalination and domestication salinity of the juvenile green crabs; carrying out numerical optimization calculation by taking the daily decreasing amplitude of salinity and the stage stable time as decision variables and the maximization of the survival rate of the fries as an optimization target to obtain an optimal combination; and determining an optimal target salinity set value of each desalination stage according to the physiological tolerance interval and an optimization result, and constructing a gradient desalination regulation and control strategy. By adopting the scheme of the application, the optimization of the salinity regulation model can be realized, and the stability and the efficiency of desalination and domestication of the young green crabs are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blue crab brackish water seed desalination domestication, more specifically, the present application relates to a blue crab brackish water seed desalination domestication interactive training system, a salinity regulation model optimization method, equipment and medium. BACKGROUND

[0002] In the blue crab breeding industry, seed desalination domestication, as a key link of brackish water breeding, is of great significance to the adaptive transfer of blue crab from seawater to low salinity environment, and to the improvement of survival rate and growth performance. Traditional desalination domestication methods rely on empirical salinity regulation, such as gradually reducing salinity and observing seed response, but lack quantitative analysis and optimization support for physiological stress mechanisms, making it difficult to cope with the variability of brackish water environment and individual differences of seed, resulting in large fluctuations in seed survival rate and unstable molting period. With the expansion of brackish water breeding scale and the increasing demand for fine management, intelligent salinity regulation model has gradually become a core technical requirement to improve seed domestication efficiency.

[0003] The existing mainstream blue crab seed desalination domestication method generally has the problems of staticity and experience dependence in model design, which is difficult to adapt to the dynamic influence of salinity change on seed osmotic pressure regulation, ion balance and physiological tolerance. Especially in the scenario of excessive daily salinity reduction or insufficient stage stabilization time, seed survival rate is easily affected by stress accumulation and significantly decreased. The root cause of this defect is that the traditional method fails to introduce a dynamic model mechanism that reflects the nonlinear correlation between salinity regulation and physiological response, lacks sensitivity analysis and optimization strategy for key parameters, and thus the regulation scheme is difficult to adapt to environmental changes and cannot fully exert the maximum potential to improve seed survival rate. Therefore, how to optimize the salinity regulation model to improve the stability and efficiency of blue crab seed desalination domestication has become a difficult problem in the industry. SUMMARY

[0004] The present application provides a blue crab brackish water seed desalination domestication interactive training system, a salinity regulation model optimization method, equipment and medium, which can optimize the salinity regulation model and improve the stability and efficiency of blue crab seed desalination domestication.

[0005] In the first aspect, the present application provides a salinity regulation model optimization method for blue crab seed desalination domestication, which comprises the following steps: Establishing a blue crab seed salinity stress response model, based on the observed values of seed survival rate in actual breeding data, performing parameter estimation and calibration on the salinity stress response model, and establishing a calibrated dynamic physiological response model; According to the dynamic physiological response model, a global sensitivity analysis is performed on the daily salinity reduction amplitude, the stage stable time and the initial salinity, and according to the global sensitivity analysis result, a physiological tolerance interval of a key regulating factor related to the salinity of the mud crab fry desalination acclimation is determined; Taking the daily salinity reduction amplitude and the stage stable time as decision variables and maximizing the survival rate of the fry as an optimization objective, numerical optimization calculation of the decision variables is performed based on the dynamic physiological response model, and an optimal combination of the daily salinity reduction amplitude and the stable time of each stage and the corresponding expected maximum survival rate are obtained. According to the physiological tolerance interval of the key regulating factor and the optimal daily salinity reduction amplitude and stable time obtained by optimization, optimal target salinity setting values of each desalination stage are determined, and a gradient desalination regulation strategy is constructed according to the optimal target salinity setting values and the regulation objective of the expected maximum survival rate.

[0006] Preferably, the mud crab fry salinity stress response model specifically includes: The survival rate data, physiological index data, ion concentration data and molting cycle data of the mud crab fry under different salinity gradients are collected; The influence coefficient of the salinity change rate on the osmoregulation and ion balance of the fry is calculated according to the survival rate data, physiological index data, ion concentration data and molting cycle data; The mud crab fry salinity stress response model, i.e. a nonlinear dynamic response model between salinity change, ion concentration and fry survival rate, is constructed through the influence coefficient.

[0007] Preferably, based on the observed values of the survival rate of the fry in the actual breeding data, parameter estimation and calibration are performed on the salinity stress response model, and the calibrated dynamic physiological response model specifically includes: The time series observation sequences of the survival rate, ion concentration and molting rate of the fry in multiple desalination cycles are extracted from the actual breeding records; The model parameters of the salinity stress response model are estimated based on the time series observation sequences; The model structure of the salinity stress response model is adjusted according to the estimated parameters, and the calibrated and optimized dynamic physiological response model is established.

[0008] Preferably, according to the dynamic physiological response model, a global sensitivity analysis is performed on the daily salinity reduction amplitude, the stage stable time, the initial salinity, the ion concentration, the temperature and the dissolved oxygen, and according to the global sensitivity analysis result, the physiological tolerance interval of the key regulating factor related to the salinity is determined specifically as follows: The parameter input ranges of the daily salinity reduction amplitude, the stage stable time, the initial salinity, the ion concentration, the temperature and the dissolved oxygen are defined in the dynamic physiological response model; The main effect index and the total effect index of each input parameter on the survival rate output of the fry are calculated. Screening sensitive parameters according to the main effect index and the total effect index, and determining the physiological tolerance interval of the key regulatory factor related to salinity based on the variation range of the sensitive parameters.

[0009] Preferably, taking the daily salinity reduction, the stage stable time and the ion concentration as the decision variables, maximizing the survival rate and the molting rate of the seedlings as the optimization objective, performing numerical optimization calculation of the decision variables based on the dynamic physiological response model to obtain the optimal combination of the daily salinity reduction, the stage stable time and the ion concentration in each stage and the corresponding expected maximum survival rate, specifically comprising: Setting the numerical search boundary of the daily salinity reduction, the stage stable time and the ion concentration; Performing multi-objective iterative solution of the decision variables within the numerical search boundary; Extracting the optimal values of the daily salinity reduction, the stage stable time and the ion concentration in each dilution stage according to the solution result, and simulating and calculating the maximum survival rate of the seedlings corresponding to the optimal values.

[0010] Preferably, according to the physiological tolerance interval of the key regulatory factor and the optimal daily salinity reduction and stable time obtained by optimization, determining the optimal target salinity setting value and ion concentration setting value in each dilution stage, specifically comprising: Determining the starting salinity in each dilution stage, taking the initial breeding salinity as the starting salinity in the first dilution stage, and taking the target salinity in the previous stage as the starting salinity in the subsequent dilution stage; Multiplying the optimal daily salinity reduction obtained by optimization by the stable time in the stage to obtain the total amount of salinity that can be reduced in the stage; Subtracting the total amount of salinity from the starting salinity in the stage, if the result is within the physiological tolerance interval of the key regulatory factor, the result is the optimal target salinity setting value in the stage.

[0011] Preferably, the salinity stress response model is a semi-mechanism semi-experience model or a mechanism model.

[0012] In the second aspect, the application provides a mangrove crab salt and alkali seedling dilution and domestication interactive training system, comprising a salinity control model optimization unit, and the salinity control model optimization unit specifically comprises: A modeling module is configured to establish a salinity stress response model of mangrove crab seedlings, perform parameter estimation and calibration on the salinity stress response model based on the observed values of the survival rate of seedlings in actual breeding data, and establish a calibrated dynamic physiological response model. A processing module is configured to perform global sensitivity analysis on the daily salinity reduction, the stage stable time, the initial salinity, the ion concentration, the temperature and the dissolved oxygen according to the dynamic physiological response model, and determine the physiological tolerance interval of the key regulatory factor related to the salinity of the mangrove crab seedling dilution and domestication according to the global sensitivity analysis result. The processing module is further configured to take the salinity daily reduction amplitude, the stage stable time and the ion concentration as decision variables, maximize the seedling survival rate and the molting rate, perform numerical optimization calculation of the decision variables based on the dynamic physiological response model, and obtain the optimal combination of the salinity daily reduction amplitude, the stable time and the ion concentration in each stage and the corresponding expected maximum survival rate. The execution module is configured to determine the optimal target salinity setting value and the ion concentration setting value in each desalination stage according to the physiological tolerance interval of the key control factor and the optimal salinity daily reduction amplitude, the stable time and the ion concentration, and construct a gradient desalination control strategy according to the optimal target salinity setting value, the ion concentration setting value, the expected maximum survival rate and the control target of the nutritional intervention.

[0013] In a third aspect, the present application provides a computer device, which comprises a memory and a processor, the memory stores a code, and the processor is configured to acquire the code and execute the salinity control model optimization method for the desalination and domestication of the blue crab seedlings.

[0014] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the salinity control model optimization method for the desalination and domestication of the blue crab seedlings.

[0015] The technical scheme provided by the embodiments of the present application has the following beneficial effects: In the embodiments of the present application, a salinity stress response model of the blue crab seedlings is established, the model is a dynamic response relationship model between the change of the salinity of the breeding water and the survival rate of the blue crab seedlings; based on the observed value of the survival rate of the seedlings in the actual breeding data, parameter estimation and calibration are performed on the salinity stress response model to establish a calibrated dynamic physiological response model; according to the dynamic physiological response model, global sensitivity analysis is performed on the salinity daily reduction amplitude, the stage stable time and the initial salinity, and according to the global sensitivity analysis result, the physiological tolerance interval of the key control factor related to the salinity of the blue crab seedlings is determined; the salinity daily reduction amplitude and the stage stable time are taken as decision variables, and the maximum survival rate of the seedlings is taken as an optimization target, numerical optimization calculation of the decision variables is performed based on the dynamic physiological response model, and the optimal combination of the salinity daily reduction amplitude and the stable time in each stage and the corresponding expected maximum survival rate are obtained; according to the physiological tolerance interval of the key control factor and the optimal salinity daily reduction amplitude and stable time, the optimal target salinity setting value in each desalination stage is determined, and a gradient desalination control strategy is constructed according to the optimal target salinity setting value and the control target of the expected maximum survival rate.

[0016] Therefore, this application obtains the optimal combination by using the daily salinity decrease and the period of stable salinity as decision variables through numerical optimization calculations, and constructs a gradient desalination regulation strategy based on the physiological tolerance range. Firstly, by establishing a salinity stress response model for mud crab seedlings, the dynamic impact of salinity changes on survival rate can be revealed based on physiological mechanisms, thus establishing a biologically based regulatory framework. The model construction process integrates survival rate, physiological indicators, and ion data to achieve a quantitative expression of the stress response, and parameter calibration is performed on this basis, enhancing the model's adaptability to actual aquaculture data from the source. Secondly, through global sensitivity analysis, the physiological tolerance range of key regulatory factors related to salinity in the desalination domestication of mud crab seedlings is determined, which can identify the parameter range that significantly affects survival rate and effectively avoid... By eliminating the blindness of the control-free strategy, this approach prioritizes the optimization of daily salinity reduction and stabilization time, enabling targeted screening of control factors. Then, with the goal of maximizing survival rate, decision variables are optimized using numerical calculation methods to solve for the optimal combination of daily salinity reduction and stabilization time, thus outputting control parameters with quantitative support. This significantly improves the efficiency of the desalination process and the expected survival rate. Finally, based on the tolerance range and optimization results, a target salinity setpoint is determined, and a gradient control strategy is constructed. This allows the control scheme to further strengthen the physiological adaptation mechanism during multi-stage desalination, minimizing the negative impact of accumulated stress on seedlings. This mechanism, through dynamic optimization, constructs an adaptive path for salinity control in saline-alkali land domestication scenarios, effectively overcoming the problem of unstable survival rates caused by experience dependence in traditional methods. In summary, this application's scheme can optimize the salinity control model, thereby improving the stability and efficiency of desalination domestication of mud crab seedlings. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a salinity control model optimization method for the desalination and domestication of mud crab seedlings, according to some embodiments of this application. Figure 2 This is a schematic flowchart illustrating the process of establishing a calibrated dynamic physiological response model according to some embodiments of this application; Figure 3 This is a flowchart illustrating the numerical optimization calculation of decision variables according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of the salinity regulation model optimization unit shown in some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a salinity control model optimization method for the desalination and domestication of mud crab seedlings, according to some embodiments of this application. Detailed Implementation

[0018] For better understanding of the technical solutions of the present application, the technical solutions of the present application will be described in detail below in combination with the drawings of the specification and specific embodiments.

[0019] Referring to Figure 1 The figure is an exemplary flow chart of a salinity regulation model optimization method for green crab fry desalination domestication according to some embodiments of the present application, which mainly includes the following steps: In step 101, a green crab fry salinity stress response model is established, and based on the observed values of fry survival rate in actual cultivation data, parameter estimation and calibration are performed on the salinity stress response model to establish a calibrated dynamic physiological response model.

[0020] It should be noted that the green crab fry salinity stress response model in the present application is a dynamic model based on the physiological mechanism of green crab fry, which can be implemented using a semi-mechanism semi-experience model or a mechanism model.

[0021] In some embodiments, the establishment of the green crab fry salinity stress response model can be achieved in the following way, that is, collecting survival rate data, physiological index data (such as osmotic pressure, enzyme activity), ion concentration data (such as Na+, K+ concentration) and molting cycle data of green crab fry under different salinity gradients, and synchronously recording temperature and dissolved oxygen time series data; calculating the influence coefficient of salinity change rate on fry osmotic pressure regulation and ion balance and the interactive effect of temperature-salinity and dissolved oxygen-salinity according to the survival rate data, physiological index data, ion concentration data and molting cycle data; constructing the green crab fry salinity stress response model through the influence coefficient and the interactive effect, that is, a nonlinear dynamic response model between salinity change, ion concentration, temperature, dissolved oxygen and fry survival rate.

[0022] Among them, the temperature time series data is the time series data formed by continuously recording the water temperature data of the cultivation in the time dimension during the desalination domestication process of the green crab fry, and the dissolved oxygen time series data is the time series data formed by continuously recording the dissolved oxygen content data of the cultivation in the time dimension during the desalination domestication process of the green crab fry, which will not be described here.

[0023] It should be noted that the calculation of the influence coefficient of salinity change rate on fry osmotic pressure regulation and ion balance and the temperature-salinity interactive effect coefficient and the dissolved oxygen-salinity interactive effect coefficient according to the survival rate data, physiological index data, ion concentration data and molting cycle data can be achieved in the following way: The collected survival rate, physiological index, ion concentration and molting cycle data under different salinity gradients are classified according to the salinity change rate to obtain multiple groups of data corresponding to different salinity change rates, for example, grouping according to different rates of 0.5 ppt, 1 ppt, 1.5 ppt, 2 ppt per day, ensuring that each group of data corresponds to only one salinity change rate, and analyzing the influence of a single factor of salinity, and then analyzing the interaction of temperature, dissolved oxygen and salinity; then, for the data corresponding to each salinity change rate, record the osmotic pressure change of the seedlings (for example, whether the osmotic pressure is maintained in the stable interval, the amplitude of exceeding the stable interval) and the ion concentration change (for example, whether the Na+ and K+ concentrations are unbalanced, and the numerical range of the imbalance), and at the same time, combined with the molting cycle (for example, whether the molting cycle is extended when the rate is too fast) and the survival rate (for example, whether the survival rate decreases when the rate is too high), the direct correlation between the salinity change rate and the osmotic pressure regulation ability and the ion balance state is determined; further, from the data corresponding to each salinity change rate, the core information that can directly reflect the influence of the salinity change rate on the osmotic pressure / ion balance is selected, for example, in the 1 ppt per day group, the records of the osmotic pressure being stable at 200-220 mOsm / kg and the Na+ concentration fluctuating by no more than 5 mmol / L are retained, which is only an example for illustration and does not limit the present application; further, by comparing the core information of different rate groups, the change amplitude of the osmotic pressure regulation ability and the ion balance state is calculated when the salinity change rate changes by one unit (such as 0.5 ppt / day), and the change amplitude is further quantified to obtain the change amplitude of the osmotic pressure regulation ability and the ion balance state when the salinity change rate changes by one unit, the temperature changes by 1°C, and the dissolved oxygen changes by 1 mg / L, and the influence coefficient and the temperature-salinity interaction effect coefficient and the dissolved oxygen-salinity interaction effect coefficient are obtained by further quantifying the change amplitude, and in specific implementation, the idea of fixed condition measurement, difference calculation and coefficient determination can be used, for example: Fixing the temperature and dissolved oxygen, only changing the salinity drop, testing the physiological index change of the seedlings: Salinity drops by 0.5 ppt / day, Na+ rises by 4 mmol / L; Salinity drops by 1.0 ppt / day, Na+ rises by 6 mmol / L; Salinity drops by 1.5 ppt / day, Na+ rises by 9 mmol / L; Calculate the average change of Na+ corresponding to each 1 ppt salinity drop, and the difference is the influence coefficient of salinity on ion balance, and the linear calculation result is used as the initial fitting basis of the low stress segment of the piecewise function.

[0024] The influence coefficient is calculated by using a segmented function: for example, low stress segment (0.5≤daily drop≤1.2 ppt): effect coefficient=5mmol / L·ppt (linear fitting); high stress segment (1.2<daily drop≤1.5 ppt): effect coefficient=20×daily drop-19 (nonlinear fitting, such as 1.5 ppt: 20×1.5-19=11mmol / L·ppt); forbidden segment (daily drop>1.5 ppt): effect coefficient=+∞ (considered as infeasible and directly excluded); cumulative stress effect is added: if the high stress segment is continuously present for 2 days, the effect coefficient is increased by 20% (to simulate the superimposed damage of continuous stress), that is, the physiological regulation ability of the blue crab fry under different salinity stress intensities is matched, the biological response law of “compensation - decompensation - fatal” is matched by using the segmented function to match the physiological regulation ability, and the superimposed damage of continuous stress is simulated by using the cumulative stress effect, wherein the segmented function is used to calculate the influence coefficient, the physiological regulation ability is matched, the cumulative stress effect is used to simulate the superimposed damage of continuous stress, and here, only an example is given, and the specific limitation of the present application is not limited.

[0025] Similarly, for the temperature-salinity interaction effect coefficient, for example: The salinity drop is kept at 1.0 ppt / day, and the dissolved oxygen is kept at 5 mg / L, and only the temperature is changed: When the temperature is 25℃, the Na⁺ rises by 6mmol / L (salinity alone); When the temperature is 30℃, the measured Na⁺ rises by 7mmol / L (salinity+high temperature combined effect); The difference between the two (7-6=1mmol / L) is the temperature-salinity interaction effect coefficient (representing that when the temperature rises by 5℃, the ion imbalance effect of salinity is additionally enhanced by 1mmol / L).

[0026] Similarly, the dissolved oxygen-salinity interaction effect coefficient, for example: The salinity drop is kept at 1.0 ppt / day, and the temperature is kept at 25℃, and only the dissolved oxygen is changed: When the dissolved oxygen is 5mg / L, the Na⁺ rises by 6mmol / L (salinity alone); When the dissolved oxygen is 3mg / L, the measured Na⁺ rises by 8mmol / L (salinity+low oxygen combined effect); The difference between the two (8-6=2mmol / L) is the dissolved oxygen-salinity interaction effect coefficient (representing that when the dissolved oxygen decreases by 2mg / L, the ion imbalance effect of salinity is additionally enhanced by 2mmol / L).

[0027] It should be noted that the obtained influence coefficient and interaction effect coefficient need to be verified with the actual situation until the influence coefficient and interaction effect coefficient match the actual situation, which will not be repeated here.

[0028] The salinity stress response model of the mud crab seedlings can be constructed using the aforementioned influence coefficients and interaction effect coefficients in the following manner: First, by using the influence coefficient and interaction effect coefficient, a dynamic calculation rule for ion concentration is established, namely: The daily calculated change in ion concentration consists of the sum of three parts: Baseline salinity effect: Multiply the daily salinity decrease by the salinity influence coefficient.

[0029] Temperature interaction effect: The difference between the current day's temperature and the reference temperature is multiplied by the temperature-salinity interaction effect coefficient.

[0030] Dissolved oxygen interaction effect: The difference between the baseline dissolved oxygen and the dissolved oxygen of the day is multiplied by the dissolved oxygen-salinity interaction effect coefficient.

[0031] The ion concentration at the end of the day is equal to the concentration of the previous day plus the calculated change in ion concentration for the day.

[0032] Enzyme activity correlation: Na⁺-K⁺-ATPase activity thresholds were added to the model (e.g., when the daily decrease was ≤1.5 ppt, the enzyme activity remained above 80%, with an effect coefficient of 11 mmol / L·ppt; when the daily decrease was >1.5 ppt, the enzyme activity dropped to 50%, with an effect coefficient of 5 mmol / L·ppt) to reflect the nonlinear physiological response that higher enzyme activity leads to stronger ion uptake capacity.

[0033] Secondly, a mapping relationship between ion imbalance and seedling survival rate is established, that is, based on historical observation data, a nonlinear mapping relationship between the degree of ion concentration imbalance and seedling survival rate is established. For example, firstly, extract matching data of ion concentration (e.g., Na⁺) and corresponding seedling survival rate from historical observation data, excluding interference data from abnormal fluctuations in temperature and salinity (ensuring that only the impact of ion imbalance is reflected); define the degree of imbalance, that is, first determine the normal physiological range of ions, and then calculate the ion concentration deviation value for each data point to quantify the degree of imbalance; then group and statistically analyze the trend, grouping according to the size of the deviation value, for example, slight deviation within ±10 mmol / L, moderate deviation ±10-20 mmol / L, and severe deviation above ±20 mmol / L, and calculate the average survival rate of each group to initially identify the nonlinear trend that the more severe the imbalance, the faster the survival rate decreases; finally, nonlinear fitting: use mathematical methods (e.g., Logistic curve, piecewise function) to fit the grouped data of imbalance degree - average survival rate into a calculable relationship (e.g., curve), fixing the pattern of slow decrease for slight deviation, sharp decrease for moderate deviation, and near zero for severe deviation. Finally, use historical data that was not involved in the fitting to verify the fitting results. If the survival rate prediction error is < preset threshold (e.g., 5%), the mapping relationship is effective; otherwise, fine-tune the grouping or fitting method, which will not be elaborated here.

[0034] Again, the above ion dynamic calculation rules are integrated with the mapping relationship between the determined ion concentration imbalance degree and the seedling survival rate to construct a complete dynamic response model, i.e., a nonlinear dynamic response model between salinity change, ion concentration, temperature, dissolved oxygen and seedling survival rate.

[0035] In specific implementation, the model can be verified using actual breeding data not involved in modeling, the daily environmental data is input into the model to obtain a predicted survival rate curve, and the predicted survival rate curve is compared with an actually observed survival rate curve, the model is fine-tuned and calibrated according to the deviation of the comparison result, and the model is adjusted until the predicted result of the model is highly consistent with the actual data, which will not be repeated here.

[0036] In some embodiments, based on the observed value of the seedling survival rate in the actual breeding data, the salinity stress response model is parameter estimated and calibrated to establish a calibrated dynamic physiological response model.

[0037] In some embodiments, referring to Figure 2 , based on the observed value of the seedling survival rate in the actual breeding data, the salinity stress response model is parameter estimated and calibrated to establish a calibrated dynamic physiological response model, which can be implemented in the following manner: In step 1011, time series observation sequences of seedling survival rate, ion concentration and molting rate in a plurality of desalination periods are extracted from actual breeding records, wherein the time series observation sequence is a sequence of key data related to salinity stress recorded continuously in time in the desalination and acclimation of mangrove crab seedlings, for example, a seedling survival rate data sequence, an ion concentration data sequence, which is not specifically limited here; In step 1012, the model parameters of the salinity stress response model are estimated based on the time series observation sequences, which can be implemented in the following manner, i.e.: First, determine the parameters to be estimated, for example, the impact coefficient or the ion imbalance threshold, which is not specifically limited here; Secondly, according to the physiological characteristics of the blue crab and the statistical characteristics of the time series observation sequence, the initial value of the parameter to be estimated is set, which is used as the starting benchmark for subsequent iterative optimization. In specific implementation, for example, the normal range of Na+ concentration is 180-220 mmol / L when the ion balance of the blue crab fry is known, and physiological disorder is prone to occur when the daily salinity drop is more than 2 ppt, which is used as the physiological boundary for parameter setting. Then, the statistical characteristics of the time series observation sequence are extracted, that is, the mean value of the salinity change rate (for example, the daily average drop of 1.2 ppt), the average value of the ion concentration (Na+) (for example, 205 mmol / L), and the correlation trend between the two (for example, for every 1 ppt drop in salinity, the average Na+ rises by 8-10 mmol / L) are calculated from the time series observation sequence data. The initial value of the parameter is set in combination with the two, for example, the ion imbalance threshold is set to 1.5 ppt, referring to the physiological tolerance threshold (2 ppt) and the 90% quantile of the observed salinity drop (1.8 ppt); and the "salinity-ion concentration influence coefficient" is set to 9 mmol / L·ppt according to the observed correlation trend, so as to ensure that the initial value of the parameter conforms to the physiological law and is consistent with the characteristics of the time series observation sequence data. Finally, the parameter value of the parameter to be estimated is adjusted step by step through iteration to minimize the residual sum of squares of the model prediction value and the actual observation value of the time series observation sequence, until the iteration converges, and the final parameter estimation value is obtained. Preferably, the weighted least squares method can be used for iteration, and other iterative algorithms can also be used in practice, which is not limited here, that is, the initial value of the parameter is taken as the starting point, and the value of the parameter to be estimated is adjusted step by step through iteration. The residual sum of squares between the prediction value of the model based on the current parameter and the actual observation value in the time series observation sequence is continuously calculated during the process, and the parameter is optimized with the core target of minimizing the residual sum of squares. When the iteration meets the preset convergence condition, the iteration is stopped, and the parameter value obtained at this time is the final parameter estimation value, which is not described here.

[0038] It should be noted that when the weighted least squares method is used for iteration, the weight can be set according to the reciprocal square of the error rate of each index, the error rate of the survival rate is 3%, the weight is 11, the error rate of the ion concentration is 4%, the weight is 6, and the error rate of the molting rate is 10%, the weight is 1. After calibration, the normality of the residual can be verified by Shapiro-Wilk test, and the heteroscedasticity can be eliminated by White test, which is not described here.

[0039] In step 1013, the model structure of the salinity stress response model is adjusted according to the parameter estimation value, and a calibrated dynamic physiological response model is established.

[0040] In some embodiments, adjusting the model structure of the salinity stress response model according to the parameter estimation value can be based on a statistical test method. In a specific implementation, first, the variable to be tested is determined, that is, the variable to be tested is selected from the input parameters of the salinity stress response model. For example, the input parameters include daily salinity drop, stage stabilization time, ion concentration, temperature, and dissolved oxygen. The temperature can be selected as the variable to be tested to determine whether the temperature has a significant impact on the survival rate of the seedlings. Second, a statistical test method is selected. The statistical test method can have multiple ways for different parameters and the same parameter. For example, for the temperature parameter, the statistical test method can be a parameter significance test method or a likelihood ratio test method. For example, for the parameter significance test method, it is determined whether a single parameter is significant. If the parameter is significant, the parameter value is retained. If the parameter is not significant, the parameter value is deleted. Finally, the model structure is adjusted according to the test result. For example, if a parameter is not significantly affected by the test, the parameter is deleted from the salinity stress response model. If there is multicollinearity among multiple variables, the variable that has a more significant impact on the survival rate is retained to simplify the model structure.

[0041] It should be noted that for the adjusted model, the matching degree between the predicted value of the adjusted model and the actual observation sequence during cultivation needs to be tested. If the matching degree meets the standard (i.e., the error is less than the preset threshold), the final model structure is determined. If the matching degree does not meet the standard, the model is adjusted again. For example, the daily salinity drop of 1.2 ppt, the Na⁺ concentration of 205 mmol / L, and the actual survival rate of 93% on the first day are substituted into the adjusted dynamic physiological response model to calculate the predicted survival rate of 92.3%, with an error of only 0.7%. The actual survival rate on the 15th day is 87%, and the model predicts 86.7%, with an error of 0.3%. The errors are both less than the preset threshold (2%), and thus the dynamic physiological response model after calibration and optimization can be used as the final model.

[0042] In step 102, global sensitivity analysis is performed on the daily salinity drop, stage stabilization time, and initial salinity according to the dynamic physiological response model. According to the global sensitivity analysis result, the physiological tolerance interval of the key control factor related to the salinity of the mud crab seedling desalination acclimation is determined.

[0043] In some embodiments, based on the dynamic physiological response model, a global sensitivity analysis is performed on the daily salinity decrease, stage stabilization time, initial salinity, ion concentration, temperature, and dissolved oxygen. Based on the results of the global sensitivity analysis, the physiological tolerance range of key salinity-related regulatory factors can be determined in the following manner: Define the parameter input ranges for daily salinity decrease, stage stabilization time, initial salinity, ion concentration, temperature, and dissolved oxygen in the dynamic physiological response model; first, calculate the main effect / total effect for controllable regulatory factors (e.g., daily salinity decrease, stage stabilization time, and ion concentration setpoints); and then screen for main effects > 0.5 and total effects > 0.5. A parameter of 0.7 is used as a key regulatory factor. Robustness indices (e.g., the coefficient of variation of survival rate when temperature fluctuates between 20-30℃) are then calculated for uncontrollable environmental factors (temperature, dissolved oxygen). If the robustness index is >10%, an environmental compensation term is added to the optimization (e.g., for every 1℃ increase in temperature, the daily salinity decrease is reduced by 0.1 ppt). The main effect index and total effect index of each input parameter on the seedling survival rate are calculated. Sensitive parameters are screened based on the main effect index and total effect index, and the physiological tolerance range of salinity-related key regulatory factors is determined based on the variation range of the sensitive parameters.

[0044] It should be noted that the main effect index reflects the degree of influence of a single parameter changing independently on the seedling survival rate output. That is, the main effect index needs to exclude the interaction of other parameters. Taking the daily salinity decrease as an example, the influence is represented by the change in seedling survival rate, while other parameters are fixed. Within a certain period, a low first daily salinity decrease and a high second daily salinity decrease are taken. The first and second daily salinity decreases are input into the dynamic physiological response model to obtain the corresponding seedling survival rates: the seedling survival rate is 95% when the daily salinity decrease is 1, and the seedling survival rate is 75% when the daily salinity decrease is 20% from low to high. This shows that the independent influence of this parameter is strong. The main effect index of this parameter can be simplified to 0.9, where the closer it is to 1, the greater the independent influence of this parameter. This is only an example and is not a specific limitation.

[0045] In addition, the total effect index in this application is the combined effect of the independent change of a single parameter plus the interaction between that parameter and other parameters on the survival rate. Taking the daily decrease in temperature and salinity as an example, with the temperature set at 20℃ (low) and 30℃ (high), and other parameters remaining fixed, four groups of survival rates were calculated: when the salinity was 0.5ppt + 20℃, the seedling survival rate was 94%; when the salinity was 0.5ppt + 30℃, the seedling survival rate was 96%, that is, the survival rate increased slightly with higher temperature, reflecting the interaction; when the daily decrease in salinity was 2ppt + 20℃, the seedling survival rate was 70%; when the salinity was 2ppt + 30℃, the seedling survival rate was 78%, that is, the survival rate still increased slightly with higher temperature, and the interaction was obvious.

[0046] The total effect index is calculated: the daily salinity reduction from low to high, plus the temperature interaction, the maximum survival rate of seedlings is reduced by 26% (96%-70%), which is greater than the influence of the daily salinity reduction alone (reduced by 20%), indicating that the comprehensive influence is strong, and the total effect index is simplified as 0.95, wherein the closer to 1, the greater the comprehensive influence.

[0047] It should be noted that the quantification standard of main effect index and total effect index is determined according to the relative contribution of the parameter to the output of the survival rate of seedlings, which is not specifically limited here, and the key control factor needs to meet the controllability and main effect significantly, and the environmental factor is excluded.

[0048] In addition, the sensitive parameters are screened according to the main effect index and the total effect index, which is the standard that the higher the index, the more significant the influence on the survival rate, based on the main effect index and the total effect index of each input parameter, wherein the total effect index is the main one, and the main effect index is the auxiliary one, and the parameters with high total effect index and main effect index are classified as sensitive parameters.

[0049] The physiological tolerance interval of the key control factor related to salinity is determined based on the variation range of the sensitive parameters, which is specifically as follows: The variation range of the sensitive parameters screened in the dynamic physiological response model is determined, for example, the daily salinity reduction is 0.5-2 ppt / day, and the stage stabilization time is 3-7 days; within the variation range, the survival rate of seedlings corresponding to different parameter values is simulated by the model, and the parameter value segment that can maintain the survival rate at a high level is found, that is, the parameter value segment close to the expected maximum survival rate, which is used as the physiological tolerance interval of the key control factor related to salinity.

[0050] In step 103, the daily salinity reduction and the stage stabilization time are used as decision variables, and the maximum survival rate of seedlings is used as the optimization target, and the numerical optimization calculation of the decision variables is performed based on the dynamic physiological response model, to obtain the optimal combination of the daily salinity reduction and the stabilization time of each stage and the corresponding expected maximum survival rate.

[0051] In some embodiments, referring to Figure 3 The figure is a flowchart of the numerical optimization calculation of the decision variables in some embodiments of the present application, and the daily salinity reduction, the stage stabilization time and the ion concentration are used as the decision variables, and the maximum survival rate of seedlings and the molting rate are used as the optimization target, and the numerical optimization calculation of the decision variables is performed based on the dynamic physiological response model, to obtain the optimal combination of the daily salinity reduction, the stabilization time and the ion concentration of each stage and the corresponding expected maximum survival rate, which can be achieved by the following steps: In step 1031, the salinity daily reduction, the stage stable time and the numerical search boundary of ion concentration are set, wherein the numerical search boundary is the decision variable range set by the physiological tolerance interval of the key control factor and the actual scene condition of the salt-tolerant crab seedling desalination domestication, and the search boundary is set according to the seedling stage, for example, the salinity daily reduction of the eyed larva is 0.5-1.2 ppt / day, and the stage stable time is 5-7 days; the salinity daily reduction of the juvenile crab I is 0.8-1.5 ppt / day, and the stage stable time is 4-6 days; the salinity daily reduction of the juvenile crab II is 1.2-2.0 ppt / day, and the stage stable time is 3-5 days; the model identifies the seedling stage through the molting rate (the molting rate ≥ 30% is determined as the juvenile crab I); for example, the upper limit is determined according to the physiological tolerance interval, the osmoregulation of the salt-tolerant crab seedling is limited, if the salinity daily reduction is more than 2 ppt / day, the survival rate of the seedling will decrease sharply, so 2 ppt / day is set as the upper limit; The lower limit is determined according to the actual scene of the salt-tolerant crab seedling desalination domestication, that is, in the salt-tolerant crab breeding, if the reduction is less than 0.5 ppt / day, the desalination period will be too long, and the cost will be increased, so 0.5 ppt / day can be set as the lower limit, and finally the numerical search boundary of the salinity daily reduction is set as 0.5-2 ppt / day, the above is only used for illustration and does not limit the present application; In step 1032, the decision variables are solved iteratively in the numerical search boundary, in the specific implementation, the multi-objective is set first, for example, the objectives of high survival rate and reasonable desalination period; then the initial value is selected in the numerical search boundary, for example, the salinity daily reduction is 0.5-2 ppt / day, such as 0.5, 1, 1.5 and 2, and the corresponding seedling survival rate and desalination period are calculated by substituting the model; the seedling survival rate and desalination period corresponding to each initial value are compared, and the optimal value is reserved, for example, the salinity daily reduction of 1.5 ppt / day is good, and the iteration is continued, for example, the salinity daily reduction of 1.4 ppt / day and the salinity daily reduction of 1.6 ppt / day are iteratively calculated, until the optimal value of the double objectives is found, for example, 1.2-1.5 ppt / day is the optimal value; It should be noted that the NSGA-II algorithm can output the Pareto optimal solution set (for example, the optimal schemes including the survival rate of 95%+the desalination period of 15 days or the survival rate of 92%+the desalination period of 12 days), and the target weight can be adjusted according to the scene (for example, the survival rate weight of the seedling period is 0.7, and the molting rate weight of the growing period is 0.6, which is not described here.

[0052] In step 1033, the optimal values of the salinity daily reduction, the stable time and the ion concentration of each desalination stage are extracted according to the solving result, and the maximum value of the corresponding seedling survival rate is simulated and calculated.

[0053] In specific implementation, for example, from the multi-target iterative solution result, the salinity daily reduction that can meet the high survival rate + reasonable desalination period is screened according to the classification of pre-desalination, mid-desalination and post-desalination, for example, 0.8 ppt / day for pre-desalination, 1.2 ppt / day for mid-desalination; the stable time, for example, 5 days for pre-desalination, 4 days for mid-desalination; the ion concentration, for example, 210 mmol / L for pre-desalination, 200 mmol / L for mid-desalination, as the optimal value of each stage, the optimal value of each stage is substituted into the dynamic physiological response model, the model will simulate and calculate the corresponding seedling survival rate according to the optimal value of each stage, and the highest value of the output result of all parameter combinations is the maximum seedling survival rate, which will not be described here.

[0054] In step 104, the optimal target salinity setting value of each desalination stage is determined according to the physiological tolerance interval of the key control factor and the optimal salinity daily reduction and stable time obtained by optimization, and a gradient desalination control strategy is constructed according to the optimal target salinity setting value and the control target of the expected maximum survival rate.

[0055] In some embodiments, the optimal target salinity setting value of each desalination stage is determined according to the physiological tolerance interval of the key control factor and the optimal salinity daily reduction and stable time obtained by optimization, and in specific implementation, the starting salinity of each desalination stage is first determined, that is, the initial breeding salinity is taken as the starting point of the first desalination stage, and the starting salinity of the subsequent desalination stage is the target salinity of the previous stage. Each desalination stage is divided into an adaptation period (the first 2 days, daily reduction = 0.8 x optimal daily reduction) and a stable reduction period (the remaining days, daily reduction = 1.2 x optimal daily reduction), and a 1-day buffer period (salinity remains unchanged) is set between the two stages. The salinity reduction amount of the stage is determined, that is, the total amount of salinity that can be reduced in this stage is calculated by (adaptation period daily reduction x 2 days) + (stable reduction period daily reduction x (stable time - 2 days)) (wherein the adaptation period daily reduction = 0.8 x optimal daily reduction, and the stable reduction period daily reduction = 1.2 x optimal daily reduction).

[0056] The target salinity is checked and set, that is, the starting salinity of the stage is reduced by the total amount of salinity, and if the result is within the physiological tolerance interval of the key control factor, it is the optimal target salinity of the stage.

[0057] The gradient desalination control strategy can be constructed according to the optimal target salinity setting value and the expected maximum survival rate, which can be realized in the following way, that is, first, the target salinity gradient is determined in stages: the desalination stages are divided into pre-desalination, mid-desalination and post-desalination, and the optimal target salinity of each desalination stage is arranged in turn to form a gradually decreasing salinity gradient; Secondly, configure the core parameters of each stage, that is, match the optimized salinity daily drop and stabilization time for each salinity gradient, and adjust the ion concentration to ensure that the core parameters of each stage adapt to the target salinity of the stage; finally, verify the survival rate constraint, that is, combine the parameters of each stage into the model to confirm that each step can maintain the expected maximum survival rate, that is, form a complete gradient desalination control strategy, which will not be described here.

[0058] In this application, the key salinity control factors are first screened through global sensitivity analysis, and then quantitatively optimized by combining numerical optimization methods. This process effectively avoids the blindness of traditional empirical control, significantly improves the stability of seedling survival rate and molting rate. Compared with the static control mode used in the prior art, the present scheme effectively reduces the negative impact of stress accumulation on the physiological state of seedlings by dynamically adjusting the salinity daily drop, stage stabilization time and other decision variables, thereby improving the overall domestication efficiency and economic benefits of aquaculture, and effectively meeting the fine control needs of salt and alkali land mud crab seed desalination domestication.

[0059] On the other hand, in some embodiments, the present application provides an interactive training system for salt and alkali land mud crab seed desalination domestication, which comprises a salinity control model optimization unit, which is used to Figure 4 The figure is a structural schematic diagram of a salinity control model optimization unit according to some embodiments of the present application. The salinity control model optimization unit 400 comprises a modeling module 401, a processing module 402 and an execution module 403, which are described as follows: The modeling module 401 is mainly used to establish a mud crab seed salinity stress response model in the present application. Based on the observed values of seedling survival rate in actual aquaculture data, the salinity stress response model is parameterized and calibrated to establish a calibrated dynamic physiological response model; The processing module 402 is used to perform global sensitivity analysis on the salinity daily drop, stage stabilization time, initial salinity, ion concentration, temperature and dissolved oxygen according to the dynamic physiological response model in the present application. According to the global sensitivity analysis results, the physiological tolerance interval of the key control factors related to the salinity of mud crab seed desalination domestication is determined; In addition, the processing module 402 is also used to take the salinity daily drop, stage stabilization time and ion concentration as decision variables, maximize the seedling survival rate and molting rate as the optimization target, and perform numerical optimization calculation of the decision variables based on the dynamic physiological response model to obtain the optimal combination of the salinity daily drop, stabilization time and ion concentration of each stage and the corresponding expected maximum survival rate; The execution module 403 is mainly used for determining optimal target salinity setting values and ion concentration setting values of each desalination stage according to the physiological tolerance interval of the key regulatory factor, the optimal salinity daily drop, the stable time and the ion concentration obtained by optimization, and constructing a gradient desalination regulation strategy according to the optimal target salinity setting values, the ion concentration setting values, the expected maximum survival rate and the regulation target of nutritional intervention.

[0060] In addition, the present application also provides a computer device, which comprises a memory and a processor, the memory stores code, and the processor is configured to acquire the code and execute the above-mentioned salinity regulation model optimization method for blue crab fry desalination domestication.

[0061] In some embodiments, reference is made to Figure 5 The figure is a structural schematic diagram of a computer device for implementing the salinity regulation model optimization method for blue crab fry desalination domestication according to some embodiments of the present application. The salinity regulation model optimization method for blue crab fry desalination domestication in the above-mentioned embodiments can be implemented by the computer device shown in the figure, which comprises at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504. Figure 4

[0062] The processor 501 can be a general central processing unit (CPU) or an application specific integrated circuit (ASIC).

[0063] The communication bus 502 can be used to transmit information between the above-mentioned components.

[0064] ​The memory 503 can be a readonly memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable readonly memory (EEPROM), a compact disc readonly memory (CDROM) or other optical disk storage, a magneto-optical disk storage, a magnetic disk or other magnetic storage device, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto. The memory 503 can exist independently of the processor 501, and can be connected to the processor 501 through the communication bus 502. The memory 503 can also be integrated with the processor 501.

[0065] The memory 503 is configured to store program codes for implementing the solutions of the present application, and the processor 501 is configured to execute the program codes stored in the memory 503. The program codes can include one or more software modules. The salt concentration regulation model optimization method for blue crab fry desalination domestication in the above embodiments can be implemented by one or more software modules in the program codes of the processor 501 and the memory 503.

[0066] The communication interface 504 is configured to communicate with other devices or communication networks, such as an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc., using any transceiver-like mechanism.

[0067] In specific implementations, as an example, the computer device can include multiple processors, each of which can be a single CPU processor or a multi-CPU processor. The processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0068] The computer device described above can be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device can be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0069] In addition, the present application also provides a computer readable storage medium, the computer readable storage medium stores a computer program, the computer program is executed by a processor to realize the salinity regulation model optimization method for the saltwater culture of blue crab fry.

[0070] Although the preferred embodiments of the present application have been described, those skilled in the art who, once aware of the basic inventive concept, can make additional changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0071] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A salinity regulation model optimization method for the desalination domestication of blue crab larvae, characterized by, The method comprises the following steps: A model of the salt stress response of the blue crab fry is established, and based on the observed values of the survival rate of the fry in actual breeding data, parameter estimation and calibration are performed on the salt stress response model to establish a calibrated dynamic physiological response model; Based on the dynamic physiological response model, global sensitivity analysis is performed on the daily salt reduction amplitude, stage stable time and initial salinity, and based on the global sensitivity analysis results, the physiological tolerance interval of the key control factor related to the salinity of the blue crab fry is determined; Taking the daily salt reduction amplitude and the stage stable time as the decision variables and maximizing the survival rate of the fry as the optimization objective, numerical optimization calculation is performed on the decision variables based on the dynamic physiological response model to obtain the optimal combination of the daily salt reduction amplitude and the stable time of each stage and the corresponding expected maximum survival rate; Based on the physiological tolerance interval of the key control factor and the optimal daily salt reduction amplitude and stable time obtained by optimization, the optimal target salinity setting value of each desalination stage is determined, and a gradient desalination control strategy is constructed based on the optimal target salinity setting value and the control objective of the expected maximum survival rate.

2. The method of claim 1, wherein, The establishment of the salt stress response model of the blue crab fry specifically comprises: Collecting survival rate data, physiological index data, ion concentration data and molting cycle data of the blue crab fry under different salinity gradients; Calculating the influence coefficient of the salinity change rate on the osmoregulation and ion balance of the fry based on the survival rate data, physiological index data, ion concentration data and molting cycle data; Building the salt stress response model of the blue crab fry, i.e. a nonlinear dynamic response model between salinity change, ion concentration and fry survival rate, through the influence coefficient.

3. The method of claim 1, wherein, Based on the observed values of the survival rate of the fry in actual breeding data, parameter estimation and calibration are performed on the salt stress response model to establish a calibrated dynamic physiological response model specifically comprising: Extracting time series observation sequences of the survival rate of the fry, ion concentration and molting rate in multiple desalination cycles from actual breeding records; Estimating the model parameters of the salt stress response model based on the time series observation sequences; Adjusting the model structure of the salt stress response model based on the estimated parameters to establish a calibrated and optimized dynamic physiological response model.

4. The method of claim 1, wherein, Based on the dynamic physiological response model, global sensitivity analysis is performed on the daily salt reduction amplitude, stage stable time, initial salinity, ion concentration, temperature and dissolved oxygen, and based on the global sensitivity analysis results, the physiological tolerance interval of the key control factor related to the salinity specifically comprises: Defining the parameter input range of the daily salt reduction amplitude, stage stable time, initial salinity, ion concentration, temperature and dissolved oxygen in the dynamic physiological response model; Calculating the main effect index and total effect index of each input parameter on the survival rate output of the fry; Based on the main effect index and total effect index, sensitive parameters are screened, and based on the variation range of the sensitive parameters, the physiological tolerance interval of the key control factor related to the salinity is determined.

5. The method of claim 1 wherein, Using daily salinity decrease, stabilization time, and ion concentration as decision variables, and maximizing seedling survival rate and molting rate as optimization objectives, numerical optimization calculations of the decision variables are performed based on the dynamic physiological response model to obtain the optimal combination of daily salinity decrease, stabilization time, and ion concentration for each stage, as well as the corresponding expected maximum survival rate. Specifically, this includes: Set numerical search boundaries for daily salinity decrease, stage stability time, and ion concentration; The decision variables are solved iteratively with multiple objectives within the numerical search boundary; Based on the solution results, the optimal values ​​of daily salinity decrease, stabilization time and ion concentration for each desalination stage are extracted, and the corresponding maximum seedling survival rate is simulated and calculated.

6. The method of claim 1, wherein, Based on the physiological tolerance range of the key regulatory factors and the optimized daily salinity decrease and stabilization time, the optimal target salinity setpoints for each desalination stage are determined, specifically including: Determine the starting salinity for each desalination stage, with the initial aquaculture salinity as the starting point for the first desalination stage, and the starting salinity for subsequent desalination stages as the target salinity for the previous stage. Multiply the optimal daily salinity reduction obtained from the optimization by the stabilization time of this stage to obtain the total salinity reduction that can be achieved in this stage; Subtracting the total salinity from the initial salinity of the stage, if the result is within the physiological tolerance range of the key regulatory factors, is the optimal target salinity setting for that stage.

7. The method of claim 1 wherein, The salinity stress response model is a semi-mechanistic semi-empirical model or a mechanistic model.

8. An interactive training system for desalinization and domestication of blue crab seedlings in saline-alkali soil, comprising a salinity control model optimization unit, characterized in that, The salinity regulation model optimization unit specifically includes: The modeling module is used to establish a salinity stress response model for mud crab seedlings. Based on the observed values ​​of seedling survival rate in actual aquaculture data, the module estimates and calibrates the parameters of the salinity stress response model and establishes a calibrated dynamic physiological response model. The processing module is used to perform global sensitivity analysis on daily salinity decrease, stage stabilization time, initial salinity, ion concentration, temperature and dissolved oxygen according to the dynamic physiological response model, and determine the physiological tolerance range of key regulatory factors related to salinity in the desalination and domestication of mud crab seedlings based on the results of the global sensitivity analysis. The processing module is also used to perform numerical optimization calculations of the decision variables based on the dynamic physiological response model, with the daily salinity decrease, stage stabilization time and ion concentration as decision variables, and the seedling survival rate and molting rate as optimization objectives, to obtain the optimal combination of daily salinity decrease, stabilization time and ion concentration at each stage and the corresponding expected maximum survival rate. The execution module is used to determine the optimal target salinity setpoint and ion concentration setpoint for each desalination stage based on the physiological tolerance range of the key regulatory factors and the optimized daily salinity reduction, stabilization time and ion concentration. Based on the optimal target salinity setpoint, ion concentration setpoint, expected maximum survival rate and the regulatory target of nutritional intervention, a gradient desalination regulation strategy is constructed. 9.A computer device, comprising a memory and a processor, wherein the memory stores code, and the code comprises the following steps: The processor is configured to acquire the code and execute the salinity regulation model optimization method for freshwater domestication of mud crab seedlings as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. When the computer program is executed by the processor, it implements the salinity control model optimization method for freshwater domestication of mud crab seedlings as described in any one of claims 1 to 7.

Images