Shrimp culture income simulation method and system based on big data

Abstract

The invention discloses a shrimp culture income simulation method and system based on big data, and relates to the technical field of shrimp culture, and the method comprises the following steps: collecting multi-source monitoring data of a shrimp culture site, carrying out the continuous mapping processing of the multi-source monitoring data according to a time dimension and a space dimension, and constructing a feature mapping band with continuous time; and on the basis of the feature mapping zone with continuous time, when newly added shrimp culture monitoring data enter a shrimp culture income simulation process, synchronously calculating the change rate of the corresponding environment variable, and cutting the change rate based on a preset rate threshold to form a mutation recognition fragment representing an environment variable mutation interval. According to the method, the time continuous feature mapping zone, the balance constraint weight chain and the self-stabilization extension channel are constructed, so that mutation recognition and smooth parameter regulation and control of the shrimp breeding environment are realized, stable and continuous output of income simulation is kept under the environment change, and the prediction precision and the reliability of breeding income decision making are improved.

Description

Technical Field

[0001] This invention relates to the field of shrimp farming technology, specifically to a method and system for simulating shrimp farming profitability based on big data. Background Technology

[0002] Big data-driven shrimp farming profitability simulation refers to the process of systematically modeling and dynamically extrapolating multi-dimensional factors involved in shrimp farming, such as environment, water quality, feed input, disease control, climate change, and market prices, using multi-source data fusion and large-scale data analysis technologies. By collecting real-time monitoring data from the farming area (such as dissolved oxygen, pH, ammonia nitrogen, temperature, and salinity) and combining it with historical production records and market information, a profitability prediction model is constructed. Machine learning or statistical regression algorithms are used to identify the comprehensive impact of each factor on shrimp growth rate, survival rate, and harvest weight, thereby simulating the economic output under different management strategies or climate scenarios. This simulation process not only quantifies the input-output relationship but also provides farmers with precise decision-making basis, such as optimal stocking density, feed ratio, feeding frequency, and harvesting timing. This transforms profitability prediction from experience-based judgment to data-driven, quantifiable extrapolation, ultimately achieving intelligent, visualized, and risk-controllable management of shrimp farming.

[0003] The existing technology has the following shortcomings: In shrimp farming profitability simulation, models typically rely on incremental training mechanisms to continuously absorb the latest monitoring data to maintain the accuracy of predictions. However, when environmental factors or farming conditions change drastically in a short period, causing a significant shift in the sample distribution, model parameters are prone to sudden jumps during updates, disrupting the original equilibrium and causing the profit prediction curve to change from a steady rise to a sharp decline. Such abrupt changes not only cause a severe deviation between the model output and the actual profit trend but may also trigger a series of decision-making errors, such as blindly increasing feed input, arbitrarily adjusting stocking density, or premature harvesting. Ultimately, this leads to a comprehensive decline in farming profitability, a chaotic data feedback chain, and even the failure of the entire prediction system's adaptive stability.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for simulating shrimp farming profits based on big data, so as to solve the problems in the background art mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a shrimp farming profitability simulation method based on big data, comprising the following steps: Step 1: Collect multi-source monitoring data from shrimp farming sites, and perform continuous mapping processing on the multi-source monitoring data according to the time and space dimensions to construct a time-continuous feature mapping band, so as to transform the change trajectory of shrimp farming environmental variables in time and space into a dynamic curve expression. Step 2: Based on the time-continuous feature mapping band, when new shrimp farming monitoring data enters the shrimp farming revenue simulation process, the rate of change of the corresponding environmental variable is calculated simultaneously, and the rate of change is cut based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. Step 3: In the parameter update process of shrimp farming income simulation, a balance constraint weight chain is introduced based on the mutation identification fragment. The changes in environmental variables before and after the mutation identification fragment are included in the balance constraint weight chain for regulation, so as to realize the transition update of model parameters from the mutation state to the continuous state. Step 4: Construct a self-stabilizing extension channel around the balance constraint weight chain. By adjusting the influence ratio of mutation identification fragments and historical data in parameter updates, the shrimp farming income simulation maintains a gradual parameter update rhythm when new data is continuously input. Step 5: Based on the self-stabilizing extended channel, a dynamic profit response band is formed. The parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extended channel are introduced into the shrimp farming profit simulation output process. This ensures that the profit change trend remains structurally stable when sudden disturbances occur in the farming environment, and outputs continuous and reliable shrimp farming profit simulation results.

[0007] Preferably, the steps of performing continuous mapping processing on multi-source monitoring data according to the time and spatial dimensions to construct a time-continuous feature mapping band include: For various types of monitoring equipment deployed at shrimp farming sites, multi-source monitoring data is collected, and timestamps, spatial coordinates of collection points, equipment numbers, and collection depth information are added during the data recording stage. After data collection is completed, the multi-source monitoring data is continuously mapped according to the time dimension. The environmental variable change curves are formed by arranging the timestamps in order and interpolating to supplement the data. The time axes of different variables are then uniformly aligned. After obtaining the time-continuous characteristic curve, the spatial dimension of the multi-source monitoring data within the same time period is continuously mapped. The spatial change curve is established with the spatial coordinates of the aquaculture area as a reference, and the curves are superimposed to form a multivariate spatial distribution structure. After completing the continuous mapping of the time dimension and the continuous mapping of the spatial dimension, the continuous mapping of the time dimension and the continuous mapping of the spatial dimension are fused together, and the spatial distribution curves are stacked in the order of the time axis to construct the time continuous feature mapping band.

[0008] Preferably, the step of segmenting the rate of change based on a preset rate threshold to form a mutation identification fragment characterizing the mutation interval of an environmental variable includes: Based on the established time-continuous feature mapping band, when new shrimp farming monitoring data is input into the shrimp farming revenue simulation process, the new data is correlated in the time dimension, and its position in the time-continuous feature mapping band is determined according to the timestamp and time-series docked with adjacent historical data nodes. After completing the time series docking, the rate of change for each type of environmental variable is calculated, and the change per unit time is determined by combining the numerical difference between the new data point and the feature mapping data point of the previous moment with the time interval. After obtaining the rate of change of each environmental variable, a rate threshold is set according to the characteristics of the aquaculture environment and the physiological tolerance range of shrimp, and the rate results are cut off by the threshold to identify the environmental mutation interval. After identifying the environmental mutation interval, multiple variable data within the same time period are merged and recombined to form a mutation identification segment that includes time range, variable type, rate of change, and direction of change.

[0009] Preferably, the step of introducing a balance constraint weight chain based on the mutation identification fragment, and incorporating the changes in environmental variables before and after the mutation identification fragment into the balance constraint weight chain for regulation, includes: Before the shrimp farming profit simulation enters the parameter update stage, time period matching and variable mapping are performed based on mutation identification segments. Stable time period data before the mutation and recovery time period data after the mutation are extracted to form a variable change data set containing three continuous data segments: before the mutation, during the mutation, and after the mutation. After obtaining environmental variable change data before and after the mutation identification segment, an initial structure of balanced constraint weight chain is established. With the mutation identification segment as the core, the magnitude and direction of change of each environmental variable before, during and after the mutation are incorporated into the weight relationship to form a continuous weight chain structure. After the balance constraint weight chain is established, the change in the variable corresponding to the mutation identification fragment is interactively controlled with the weight chain, and the parameter update rate and magnitude are gradually adjusted according to the weight ratio. After completing the balance constraint weight chain adjustment, the updated parameters after balance are mapped to the time-continuous feature map.

[0010] Preferably, when the balanced constraint weight chain interactively regulates the change in the variable corresponding to the mutation identification segment, it adjusts the proportion of data in the three time periods before, during, and after the mutation in the weight chain. This allows the data in the time period before the mutation to be used to suppress parameter jumps, the data in the time period during the mutation to be used to maintain a balanced transition, and the data in the time period after the mutation to be used to guide the parameter to recover to a stable state.

[0011] Preferably, the steps of constructing a self-stabilizing extension channel around the balance constraint weight chain and adjusting the influence ratio of mutation identification fragments and historical data in parameter updates include: Under the premise that the balance constraint weight chain is formed and the mutation identification fragment is regulated, the mutation identification fragment and its corresponding historical data are classified and organized, the time interval is determined according to the weight node of the balance constraint weight chain, and the adjacent historical data sequences before and after the mutation are extracted as the balance reference interval. After forming a time-continuous input structure of mutation identification fragments and historical data, the influence relationship between the two types of data in parameter updates is preset in a linked manner, the influence range of mutation identification fragments and historical data is determined, and a dynamic weight allocation relationship is established. After establishing the linkage between mutation identification fragments and historical data, this linkage is introduced into the parameter update process, and the weight ratio between mutation identification fragments and historical data is automatically adjusted according to the time correlation of newly added shrimp farming monitoring data. After the linkage adjustment mechanism between mutation identification fragments and historical data is running stably, the self-stabilizing extension channel is continuously extended, and the influence ratio of mutation identification fragments and historical data is recorded in the time-continuous feature mapping band.

[0012] Preferably, after forming a time-continuous input structure of mutation identification fragment and historical data, when the influence relationship between the two types of data in parameter update is preset in a linked manner, the influence range of the mutation identification fragment in the early stage of parameter update covers the period of mutation occurrence and the short-term response interval, and the influence range of historical data covers the stable state before mutation and the recovery stage after mutation.

[0013] Preferably, the steps for introducing the parameter update results adjusted by the balance constraint weight chain and the self-stabilizing extended channel into the shrimp farming profit simulation output process, and outputting the shrimp farming profit simulation results, include: Under the premise of stable operation of the self-stabilizing extended channel, the parameter update results after joint adjustment by the balance constraint weight chain and the self-stabilizing extended channel are integrated in the time dimension to form a continuous update sequence that reflects the dynamic change trend of parameters, and matched with the corresponding time continuous feature mapping band. After obtaining the adjusted parameter time-continuous structure, a spatial extension structure of the dynamic benefit response band is constructed based on the output path of the self-stabilizing extension channel. The spatial distribution characteristics of different parameters are synchronously mapped with the time-continuous parameter curve as the main axis, forming a smooth transition structure across regions. After establishing the characteristics of temporal continuity and spatial extension, the resulting dynamic profit response band is introduced into the shrimp farming profit simulation output process to synchronously adjust the variables in the profit calculation, so that the profit prediction forms a stable profit change curve under the drive of continuously changing parameters. Based on the dynamic return response band participating in the return output, the continuous results of return change trend are dynamically fed back and integrated. The output data is aligned and recorded with the time continuous feature mapping band, and used as a reference input in subsequent prediction periods.

[0014] Preferably, when introducing the shrimp farming revenue simulation output process into the dynamic revenue response band, the parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extension channel are used as the revenue calculation input according to the synchronous rhythm of the time continuous feature mapping band.

[0015] The shrimp farming profit simulation system based on big data includes a multi-source monitoring data mapping module, an environmental variable mutation identification module, a balance constraint weight chain regulation module, a self-stabilizing extension channel construction module, and a dynamic profit response band generation module. The multi-source monitoring data mapping module collects multi-source monitoring data from shrimp farming sites and performs continuous mapping processing on the multi-source monitoring data according to the time and spatial dimensions to construct a time-continuous feature mapping band. The environmental variable mutation identification module, based on the time-continuous feature mapping band, simultaneously calculates the rate of change of the corresponding environmental variable when new shrimp farming monitoring data enters the shrimp farming profit simulation process, and cuts the rate of change based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. The balance constraint weight chain regulation module introduces a balance constraint weight chain based on mutation identification fragments during the parameter update process of shrimp farming profit simulation. The changes in environmental variables before and after the mutation identification fragments are incorporated into the balance constraint weight chain for regulation. The self-stabilizing extended channel construction module builds a self-stabilizing extended channel around the balance constraint weight chain, and adjusts the influence ratio of mutation identification fragments and historical data in parameter updates in a linked manner. The dynamic profit response band generation module forms a dynamic profit response band based on a self-stabilizing extended channel. It introduces the parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extended channel into the shrimp farming profit simulation output process, and outputs the shrimp farming profit simulation results.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention achieves dynamic tracking of aquaculture environmental variables in both time and space by constructing a time-continuous feature mapping band and mutation identification segment during shrimp aquaculture profit simulation. This enables the model to accurately identify environmental mutation intervals when monitoring data changes rapidly. By introducing a balanced constraint weight chain during parameter updates, the environmental changes before and after mutations are smoothly controlled, ensuring continuous transition of model parameters under sudden disturbances. This avoids fluctuations in the profit curve caused by parameter instability, thus ensuring that the profit prediction results are consistent with the actual aquaculture environment change trend and improving the continuity and reliability of the simulation results.

[0017] This invention constructs a self-stabilizing extension channel around the balance constraint weight chain and forms a dynamic benefit response band in the benefit output stage. This enables the model to adaptively adjust the parameter influence ratio when facing a continuous input data stream, thereby allowing the benefit simulation process to maintain a gradual update rhythm in long-term operation. When sudden changes occur in the aquaculture environment, the system can achieve flexible buffering through the dynamic benefit response band, keeping the benefit prediction output structurally stable, ensuring a smooth and continuous prediction trend, and improving the self-stabilization of benefit simulation and the controllability of economic decision-making. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of the method for simulating shrimp farming profits based on big data, as described in this invention.

[0020] Figure 2 This is a schematic diagram of the modules of the shrimp farming profit simulation system based on big data of the present invention. Detailed Implementation

[0021] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0022] This invention provides, for example Figure 1 The shrimp farming profitability simulation method based on big data shown includes the following steps: Step 1: Collect multi-source monitoring data from shrimp farming sites, and perform continuous mapping processing on the multi-source monitoring data according to the time and space dimensions to construct a time-continuous feature mapping band, so as to transform the change trajectory of shrimp farming environmental variables in time and space into a dynamic curve expression. The specific implementation method for this step is as follows: Multi-source monitoring data, including water quality, meteorological data, and aquaculture operation records, was collected from various types of monitoring equipment deployed at shrimp farming sites. Water quality data specifically included indicators such as water temperature, dissolved oxygen, pH, ammonia nitrogen, nitrite, salinity, transparency, oxidation-reduction potential, and suspended solids concentration. Meteorological data included wind speed, wind direction, air temperature, humidity, rainfall, light intensity, air pressure, and diurnal temperature range. Aquaculture operation data included daily feed intake, feed type, stocking density, water change frequency, aerator operating time, and harvesting time. All collected data included timestamps and spatial coordinates of the collection points. To ensure data integrity and comparability, all types of monitoring data were sampled at uniform time intervals, determined based on the scale of farming and monitoring frequency, for example, with a minimum recording unit of 5 or 10 minutes. All collected data was stored in its original format, with the monitoring equipment number, geographical coordinates, sampling depth, and data source identification information appended during the data recording stage. Through these operations, the monitoring data from the shrimp farming site possessed complete temporal and spatial dimension indexes, providing a basic data framework for subsequent continuous mapping.

[0023] After data collection, various monitoring data are continuously mapped along the time dimension. Specifically, all monitoring data are first arranged in timestamp order, and missing time segments are supplemented by interpolation using data from adjacent time points to ensure the continuity of the time series. Subsequently, the environmental variable changes at the same monitoring point at different times are correlated and calculated, transforming the original discrete sampling points into time-continuous change curves. For example, for water temperature data, a water temperature change curve is formed on a continuous time axis according to sampling intervals, allowing the rise, stabilization, and fall of water temperature to be presented in a coherent form. Variables such as dissolved oxygen, pH, and ammonia nitrogen are also processed in the same way to achieve time continuity. Through this processing, all environmental variables at each monitoring point form a smooth curve structure along the time dimension. Simultaneously, to maintain temporal synchronization between different variables, the time axes of all variables are uniformly aligned, ensuring that different types of data have a one-to-one corresponding numerical state at the same time, thus establishing a complete foundation for time series mapping. After this processing step, the dynamic changes in the shrimp farming environment can be clearly reflected in the time dimension, providing a continuous time reference for spatial mapping.

[0024] After obtaining the temporally continuous characteristic curves, a continuous spatial mapping is performed on the multi-source monitoring data within the same time period. Specifically, using the spatial coordinates of the aquaculture area as a reference, the data from each monitoring point collected at the same time are arranged according to geographical location, for example, establishing a data sequence according to pond coordinates from west to east or from south to north. For data of the same type of environmental variable, the numerical differences between adjacent monitoring points are calculated, and the spatial trend is determined based on the degree of difference. When a variable changes smoothly in space, the values ​​of adjacent monitoring points are linearly connected to form a continuous spatial variation curve; when a variable has significant spatial differences, interpolation calculations using dense sampling points are used to supplement the transition segment, thus constructing a smooth spatial variation relationship. In this way, spatial distribution curves for key variables such as water temperature, dissolved oxygen, and salinity are formed under the same time conditions. By overlaying the spatial distribution curves of different variables, a multivariate spatial distribution structure at that moment can be constructed, thus fully reflecting the environmental differences and local characteristics of the entire aquaculture area at the spatial level. This continuous spatial mapping allows the environmental state of each geographical location in the aquaculture area to be continuously represented at the same time, establishing a spatial foundation for the integration of time and space.

[0025] After completing continuous mapping in both the temporal and spatial dimensions, the two are fused to construct a temporally continuous feature mapping band. Specifically, the spatial distribution curves generated in each time segment are stacked sequentially along the time axis, creating a continuously changing dynamic structure in the temporal dimension. In this way, each monitored variable forms a continuous dynamic curve in both time and space, intuitively reflecting its spatiotemporal trajectory during the aquaculture cycle. To ensure the integrity of the mapping band, the time axes of all variables are synchronously adjusted to ensure precise correspondence between spatial distributions at the same time points. Through the bidirectional correlation between the time and spatial axes, a continuous feature band composed of both temporal changes and spatial distribution is formed, demonstrating continuous change in the aquaculture environment in both time and space. This feature mapping band comprehensively represents the dynamic changes of environmental variables in shrimp farming, reflecting the environmental evolution trend of the aquaculture cycle in the temporal dimension and showcasing the environmental differences between different farming units in the spatial dimension. Ultimately, this feature mapping band can transform multi-source monitoring data from shrimp farming sites into continuously analyzable dynamic curves, providing a unified reference basis for identifying the distribution of environmental changes in subsequent shrimp farming profitability simulations.

[0026] Through the implementation of the above steps, the multi-source monitoring data from shrimp farming sites are transformed layer by layer from their original discrete state into a comprehensive feature structure with temporal continuity and spatial correlation. The continuous mapping in the temporal dimension ensures that the dynamic changes of environmental variables during the farming cycle can be fully reflected, while the continuous mapping in the spatial dimension ensures that environmental characteristics between different monitoring points can be smoothly transitioned. The fusion of temporal and spatial processing ultimately forms a temporally continuous feature mapping band, enabling the shrimp farming environment to be dynamically expressed in both temporal and spatial dimensions.

[0027] Step 2: Based on the time-continuous feature mapping band, when new shrimp farming monitoring data enters the shrimp farming revenue simulation process, the rate of change of the corresponding environmental variable is calculated simultaneously, and the rate of change is cut based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. The specific implementation method for this step is as follows: Based on the established temporally continuous feature mapping band, when new shrimp farming monitoring data is collected and input into the shrimp farming profitability simulation process, the first step is to correlate the new monitoring data with the temporal dimension of the existing feature mapping band. Specifically, after the new data arrives, its position in the temporally continuous feature mapping band is determined based on its timestamp information, and it is then time-series aligned with adjacent historical data nodes. Through this operation, a continuous data chain consisting of historical data nodes and new data nodes can be formed on the time axis, allowing the new data to naturally connect to the original temporally continuous feature curves. Furthermore, the environmental variables to which the new data belongs are identified and classified, mapping them to specific variable curves in the feature mapping band, such as water temperature curves, dissolved oxygen curves, pH curves, and ammonia nitrogen curves. In this way, the new monitoring data is not only incorporated into the temporally continuous feature mapping structure but also gains a spatial positional reference, enabling the calculation of the rate of change within the complete spatiotemporal feature system.

[0028] After the newly added monitoring data is mapped to the time-continuous feature map, the rate of change for each type of environmental variable is calculated to reflect the dynamic trend of the variable over time. Specifically, for each variable curve, the numerical difference between the newly added data point and the feature map data point at the previous moment is selected, and the change per unit time is calculated based on the time interval. For example, for dissolved oxygen, the rate of change during that period is obtained by comparing the difference between the newly added dissolved oxygen value and the dissolved oxygen value at the previous collection moment. Similarly, the rate of change is calculated sequentially for variables such as water temperature, pH, ammonia nitrogen, and salinity. Throughout this process, it is ensured that the rate of change for each variable is referenced to a uniform time interval to maintain consistency in the time response between different variables. The calculated rate of change reflects the intensity and direction of fluctuations in environmental variables over a short period, providing a basis for judging sudden environmental changes. In this way, the originally static newly added monitoring data can be transformed into dynamic change signals, enabling it to reflect real-time changes in the aquaculture environment.

[0029] After obtaining the rate of change of each environmental variable, threshold segmentation is performed on the rate results of all variables to identify intervals that may represent environmental abrupt changes. To this end, thresholds for the rate of change of each variable are pre-set based on the environmental characteristics of the aquaculture area, the aquaculture model, and the physiological tolerance range of the shrimp. For example, a rate of change of water temperature exceeding the set threshold per unit time may indicate a sudden change in air temperature or the introduction of cold or warm water flows; a sudden increase or decrease in dissolved oxygen rate may indicate abnormal operation of aerators or eutrophication; a large change in pH within a short period may be related to ammonia nitrogen concentration or feeding behavior. By comparing the rate results of each variable with the preset thresholds, it is possible to determine which time periods have data changes exceeding the normal fluctuation range. When the rate of change of a variable exceeds its corresponding threshold, that time period is marked as an abnormal interval, and its start time, end time, variable type, and direction of change are recorded. Furthermore, if multiple environmental variables simultaneously exhibit abnormal rates within similar time periods, they are combined and identified as a composite abrupt change interval to reflect a more complex state of environmental change.

[0030] After identifying the anomalous change intervals of each environmental variable, these intervals are merged and reorganized to form complete mutation identification segments. Specifically, data from multiple variables that undergo anomalous changes within the same time period are integrated to create a continuous segment structure for the same mutation event in time. For example, if both water temperature and dissolved oxygen experience rate abrupt changes within the same time period, this time period is merged into a single mutation identification segment. This segment simultaneously records the rate and direction of change of water temperature and dissolved oxygen, serving as a comprehensive expression of the environmental mutation. For mutation intervals with long durations or strong fluctuations across multiple adjacent time periods, they are extended to form long-term mutation segments to reflect the dynamic change trend of the aquaculture environment over a certain period. The resulting mutation identification segments contain multi-dimensional information such as time range, variable type, rate of change amplitude, and direction of change, comprehensively characterizing the mutation features of shrimp aquaculture environmental variables within a specific time period. By mapping these mutation identification segments to time-continuous feature mapping bands, regions with abrupt changes in continuous environmental change curves can be clearly identified. This allows the entire shrimp aquaculture profitability simulation to quickly identify potential environmental anomalies when processing new data, providing a basis for introducing balance constraints during subsequent parameter updates.

[0031] Through the above steps, newly added shrimp farming monitoring data is effectively incorporated into the time-continuous feature mapping band structure after entering the shrimp farming profitability simulation process. By calculating the rate of change and threshold segmentation, precise identification of abrupt changes in environmental variables is achieved. This abrupt change identification segment not only reflects the rapid fluctuations of the farming environment in the time dimension but also maintains a spatial correspondence with existing feature mapping bands, thus revealing the dynamic abrupt change characteristics of the farming environment within a dual continuous context of time and space. In this way, shrimp farming profitability simulation can maintain an immediate response to environmental changes during continuous data updates, enabling the simulation process to identify and handle sudden environmental changes.

[0032] Step 3: In the parameter update process of shrimp farming income simulation, a balance constraint weight chain is introduced based on the mutation identification fragment. The changes in environmental variables before and after the mutation identification fragment are included in the balance constraint weight chain for regulation, so as to realize the transition update of model parameters from the mutation state to the continuous state. The specific implementation method for this step is as follows: Before the shrimp farming profitability simulation enters the parameter update phase, time period matching and variable mapping are performed on the mutation identification fragments generated in the previous stage. Specifically, based on the time range recorded in the mutation identification fragments, stable time period data before the mutation and recovery time period data after the mutation are extracted to form a variable change data set containing consecutive time periods before and after the mutation. In this set, each environmental variable has three consecutive data segments: before the mutation, during the mutation, and after the mutation. In this way, complete time series information corresponding to the mutation identification fragment can be obtained, and the trend and magnitude of change of each variable before and after the mutation can be clearly defined. For example, when water temperature rises rapidly in the mutation identification fragment, the magnitude and transition trend of the variable can be obtained by extracting the stable water temperature change curve before the mutation and the slow adjustment curve after the mutation; similarly, similar before-and-after time period comparison data can be formed for variables such as dissolved oxygen, pH, salinity, and ammonia nitrogen. Through the above operations, data support and variable references are provided for the subsequent introduction of a balanced constraint weight chain.

[0033] After acquiring environmental variable change data before and after the mutation identification segment, an initial structure of a balanced constraint weight chain is established. This balanced constraint weight chain uses the mutation identification segment as its core, taking the corresponding environmental variable changes before and after the mutation as input, and incorporating the magnitude and direction of each variable's changes before, during, and after the mutation into the weighting relationship. In practice, firstly, based on the types of environmental variables involved in the mutation identification segment, the proportional relationship of each variable's changes before and after the mutation is determined, and initial weights are assigned according to the variables' importance. For example, dissolved oxygen has a significant impact on shrimp metabolism, so its weight can be higher than water temperature or salinity; while water temperature changes are more sudden, its weight allocation can be biased towards the period of mutation. In this way, each environmental variable occupies a different position in the weight chain and is linked to its corresponding trend of change. This balanced constraint weight chain, by comprehensively measuring the directionality, rate, and magnitude of variable changes, forms a continuous weight chain structure that dynamically reflects the balance of environmental changes, providing a smooth constraint basis for subsequent parameter updates.

[0034] After the balance constraint weight chain is established, the variable changes corresponding to the mutation identification segment are interactively controlled with the weight chain to form a transition update mechanism from a mutation state to a continuous state during parameter updates. Specifically, during parameter updates, parameter changes within the time range of the mutation identification segment are not directly replaced in full. Instead, the parameter update rate and magnitude are gradually adjusted according to the proportion of the balance constraint weight chain, ensuring that the parameters maintain a continuous trend of change during the mutation period. For example, when the mutation identification segment reflects a rapid increase in water temperature, while the balance constraint weight chain shows a significant difference in environmental state before and after the mutation, the weight proportion of data from the period before the mutation is increased to reduce the direct impact of data from the period during the mutation, thus making the parameter update more inclined towards a smooth transition state. Conversely, when the mutation identification segment shows that environmental variables quickly recover to a stable state after the mutation, the weight proportion of the period after the mutation is increased to make the parameter update quickly conform to the new balance trend. Through this process, parameter changes are no longer directly driven by mutation data, but are gradually transitioned under the action of the balance constraint weight chain, thereby avoiding abrupt jumps in the parameter update process.

[0035] After adjusting the changes in the mutation identification fragment variables through the balance constraint weight chain, the balanced update results are continuously injected into the parameter update process of the shrimp farming profitability simulation to achieve dynamic balance of the model as a whole. Specifically, the parameter changes processed by the balance constraint are mapped to a time-continuous feature map, ensuring that the updated parameters maintain continuity with the feature map in the time dimension and maintain coordination with adjacent monitoring points in the spatial dimension. When new farming data continues to enter the simulation process, the system automatically adjusts the weight distribution of different variables according to the current state of the balance constraint weight chain, ensuring that the parameter updates always maintain a dynamic and stable rhythm. During this process, the abnormal periods represented by the mutation identification fragments are not directly excluded, but are integrated into the overall update sequence in a balanced form, thus preserving the real impact of environmental mutations while avoiding abnormal parameter jumps caused by sudden changes. In this way, the parameter update process maintains continuity in time and stability in state, ultimately achieving a smooth transition of the model from a mutation state to a continuous state.

[0036] Through the execution of the above steps, the parameter update process of shrimp farming profit simulation achieves dynamic balance control based on mutation identification fragments. By introducing a balance constraint weight chain, it is possible not only to rationally allocate the influence ratio of each environmental variable before and after the mutation when the mutation identification fragment appears, but also to achieve a natural transition from the mutation state to the continuous state during the parameter update process. This process effectively avoids instantaneous jumps in parameters caused by drastic environmental changes, ensuring that the profit simulation maintains a stable response capability when facing complex changes in the farming environment, and making the model's parameter adjustments more consistent with the gradual characteristics of environmental changes in actual farming scenarios.

[0037] Step 4: Construct a self-stabilizing extension channel around the balance constraint weight chain. By adjusting the influence ratio of mutation identification fragments and historical data in parameter updates, the shrimp farming income simulation maintains a gradual parameter update rhythm when new data is continuously input. The specific implementation method for this step is as follows: With the balance constraint weight chain already formed and the mutation identification segment regulation completed, the mutation identification segment and its corresponding historical data are categorized and organized to construct the input foundation for the self-stabilizing extension channel. Specifically, based on the weight nodes in the balance constraint weight chain, the time intervals corresponding to the mutation identification segment and the historical data are determined. The mutation identification segment mainly reflects the dynamic impact of short-term environmental changes, while the historical data reflects the fundamental laws of long-term environmental evolution and stable cycles. To ensure a reasonable weight allocation for the two types of data in subsequent coordinated regulation, the time boundaries of the mutation identification segment dataset are defined, and adjacent historical data sequences before and after the mutation are extracted as balance reference intervals. For example, when the mutation identification segment shows rapid fluctuations in water temperature and dissolved oxygen within a few hours, by selecting stable data from two days before and after this time period as historical references, a continuous time data structure containing both short-term mutation and long-term stability characteristics can be formed. In this way, the mutation identification segment and historical data are arranged in an orderly manner in a time series, providing a time-continuous input source for the subsequent construction of the self-stabilizing extension channel.

[0038] After establishing a temporally continuous input structure for the mutation identification fragment and historical data, the influence relationship between the two types of data in parameter updates is pre-set to create a balanced framework for the self-stabilizing extended channel. Specifically, the influence range of the mutation identification fragment in parameter updates is first determined, typically covering the period of the mutation and the subsequent short-term response interval. Then, the influence range of historical data is determined, spanning a longer period and encompassing the stable state before the mutation and the recovery phase afterward. Based on this, a linkage relationship is established between the two types of data, giving the mutation identification fragment a high response weight in the initial parameter updates, gradually decreasing its influence over time; simultaneously, the influence weight of historical data in parameter updates gradually increases to form a dynamic equilibrium. For example, in the initial update phase after a mutation, the weight of the mutation identification fragment may account for more than 60% to ensure the model can quickly respond to environmental changes; while after several cycles of updates, the weight of historical data gradually increases to become dominant, maintaining the stability of the parameters under the overall trend. Through this dynamic allocation of weights, a natural evolutionary process from short-term response to long-term stability can be formed in parameter updates; this process is the basic balance mechanism of the self-stabilizing extended channel.

[0039] After establishing the linkage between mutation identification fragments and historical data, this linkage is incorporated into the parameter update process, enabling the self-stabilizing extension channel to dynamically adjust. Specifically, each time new shrimp farming monitoring data enters the profit simulation process, the self-stabilizing extension channel is adjusted synchronously according to the update rhythm of the time-continuous feature mapping band. By identifying the temporal correlation between the new data and existing mutation identification fragments, it is determined whether the new data belongs to a mutation range or a stable range. If the new data belongs to a mutation range, the weight of the mutation identification fragment is increased during parameter updates, allowing the model to respond promptly to short-term environmental changes. If the new data belongs to a stable range, the weight of historical data is gradually increased, making the parameter update process more influenced by long-term trends. In this way, the rate and direction of parameter updates automatically adjust over time, preventing drastic parameter fluctuations due to mutation data and avoiding reduced response sensitivity due to over-reliance on historical data. This linkage adjustment process allows the model to maintain a gradual parameter update rhythm while continuously inputting new data, thus avoiding overshoot or lag issues caused by sudden changes.

[0040] After the linkage adjustment mechanism between mutation identification fragments and historical data stabilizes, the self-stabilizing extension channel is continuously extended to ensure long-term equilibrium over time. Specifically, at the end of each parameter update cycle, the influence ratio of mutation identification fragments and historical data in the current cycle is automatically recorded in a time-continuous feature map, serving as the initial weight distribution reference for the next cycle. When a new mutation event occurs, the system can quickly and adaptively adjust based on the existing weight distribution relationship, allowing the new mutation identification fragment to naturally integrate into the existing self-stabilizing extension channel without disrupting the previous equilibrium structure. Over time, the entire self-stabilizing extension channel forms a time-extended trajectory containing multiple historical equilibrium relationships. In this trajectory, the weight influence of different mutation events is smoothly connected, and the historical evolution trends of each variable are gradually absorbed, thus achieving the self-stabilizing characteristics of the model in long-term operation. Through this extension method, shrimp farming profitability simulation can achieve dynamic self-equilibrium during continuous data input, ensuring that the parameter update process remains continuous, gradual, and adjustable.

[0041] Through the above steps, a self-stabilizing extension channel constructed around the balanced constraint weight chain is realized, enabling a dynamically linked balance between the influence ratio of mutation identification fragments and historical data during parameter updates. This implementation method not only allows for rapid response to environmental mutations in the short term but also maintains the smoothness and stability of parameter updates during long-term operation. This allows shrimp farming profitability simulation to maintain a gradual parameter update rhythm even when faced with continuously inputting new monitoring data.

[0042] Step 5: Based on the self-stabilizing extended channel, a dynamic benefit response band is formed. The parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extended channel are introduced into the shrimp farming benefit simulation output process, so that the benefit change trend remains structurally stable when the farming environment is suddenly disturbed, and the continuous and reliable shrimp farming benefit simulation results are output. The specific implementation method for this step is as follows: Under the premise of stable operation of the self-stabilizing extended channel, the parameter update results after joint adjustment by the balanced constraint weight chain and the self-stabilizing extended channel are integrated in the time dimension to form a continuous update sequence that reflects the dynamic change trend of parameters. Specifically, according to the parameter update rhythm of the self-stabilizing extended channel in different time periods, the weight change curves of each parameter are integrated to form a smooth transition continuous change structure on the time axis. This continuous change structure includes water temperature regulation parameters, dissolved oxygen related parameters, feed feeding efficiency parameters, stocking density related parameters, and disease impact correction parameters. Through this integration method, all core parameters involved in the benefit calculation form a continuous change curve in the time dimension, thereby avoiding parameter fluctuations caused by local mutations. At the same time, these parameter change curves are matched with the time continuous feature mapping band to ensure that the parameter update results are consistent with the time dynamic characteristics of the aquaculture environment, achieving two-dimensional synchronization of parameters and environment. Through this step, a time continuous basic structure is laid for the formation of the dynamic benefit response band, enabling the benefit simulation output stage to inherit the smooth characteristics of parameter changes in the time dimension.

[0043] After obtaining the adjusted, time-continuous parameter structure, a spatial extension structure of the dynamic revenue response band is constructed based on the output path of the self-stabilizing extension channel. Specifically, the spatial distribution characteristics corresponding to different parameters are synchronously mapped using the time-continuous parameter curve as the main axis. Since shrimp farming environments have spatial distribution characteristics, parameter changes not only need to reflect temporal continuity but also spatial coordination. Therefore, when constructing the dynamic revenue response band, the parameter change results of each monitoring point or farming unit are arranged according to their geographical coordinates, and a smooth transition structure across regions is formed through spatial interpolation and continuous correlation. In this way, at the same time, the revenue prediction results between different farming units can maintain a coordinated change relationship in space. For example, when a farming area adjusts its parameters due to increased water temperature, the parameter updates in the surrounding areas will automatically adjust according to the linkage ratio of the self-stabilizing extension channel, so that the overall revenue prediction results form a continuous response band in the spatial dimension. Through this spatial extension processing, the dynamic revenue response band not only reflects the revenue change at a single point in time but also comprehensively presents the continuity of revenue distribution within the farming area, forming a dynamic response structure that spans both time and space.

[0044] After establishing the temporal continuity and spatial extension characteristics, the resulting dynamic profit response band is introduced into the output process of shrimp farming profit simulation to structure the profit prediction results. Specifically, firstly, based on the temporal change trajectory and spatial distribution of each parameter in the dynamic profit response band, the variables involved in the profit calculation process are synchronously adjusted. By coordinating the parameter change rate, fluctuation amplitude, and spatial extension range, the profit calculation maintains a balanced relationship in the input stage. Subsequently, the adjusted parameter change sequence is input into the profit change trend calculation process, enabling the profit prediction to gradually form a stable profit change curve driven by continuously changing parameters. In this process, the dynamic profit response band acts as a transition layer, continuously mapping parameter changes that may cause abrupt changes in the profit output stage, thereby avoiding sudden fluctuations in the output results. For example, when the mutation identification segment shows a sharp drop in dissolved oxygen in the farming environment, the dynamic profit response band will smooth the impact of this variable on the profit calculation based on the adjustment results of the balance constraint weight chain and the self-stabilizing extension channel, so that the final profit change trend remains continuous and interpretable. This structured output enables shrimp farming revenue simulation to maintain a stable revenue forecast trend even under sudden disturbances.

[0045] Based on the dynamic benefit response band's role in benefit output, the continuous results of benefit trend changes are dynamically fed back and integrated, forming a closed, self-stabilizing update cycle for the entire benefit simulation process. Specifically, after each benefit prediction cycle, the output data in the dynamic benefit response band is aligned and recorded with the time-continuous feature mapping band, creating a traceable continuous trajectory for the benefit prediction results over time, which serves as the reference input for the next benefit simulation cycle. When new aquaculture monitoring data arrives, the system automatically adjusts the initial equilibrium state of parameter updates based on the dynamic benefit response band formed in the previous cycle, ensuring that the new round of parameter changes continues on the existing smooth structure. In this way, the dynamic benefit response band not only plays a smoothing role in a single output cycle but also forms a continuously extending self-stabilizing structure during long-term operation, enabling shrimp aquaculture benefit simulation to maintain a consistent update rhythm and stable prediction trend under continuous input data. Therefore, when sudden disturbances occur in the aquaculture environment, the benefit prediction curve will not jump or break, but will achieve a flexible transition through the buffering effect of the dynamic benefit response band, maintaining the overall structural stability of the benefit trend.

[0046] Through the specific implementation of the above steps, the dynamic profit response band formed by the self-stabilizing extended channel achieves a smooth connection between the parameter adjustment results and the profit simulation output, enabling shrimp farming profit prediction to maintain a stable structural relationship and continuous output trend in the face of complex environmental changes. By introducing the parameter update results adjusted by the balanced constraint weight chain and the self-stabilizing extended channel into the profit output process, the dynamic profit response band continuously expresses parameter changes in both time and space dimensions, thus enabling shrimp farming profit simulation to output continuous and reliable profit results even under sudden disturbance conditions.

[0047] This invention achieves dynamic tracking of aquaculture environmental variables in both time and space by constructing a time-continuous feature mapping band and mutation identification segment during shrimp aquaculture profit simulation. This enables the model to accurately identify environmental mutation intervals when monitoring data changes rapidly. By introducing a balanced constraint weight chain during parameter updates, the environmental changes before and after mutations are smoothly controlled, ensuring continuous transition of model parameters under sudden disturbances. This avoids fluctuations in the profit curve caused by parameter instability, thus ensuring that the profit prediction results are consistent with the actual aquaculture environment change trend and improving the continuity and reliability of the simulation results.

[0048] This invention constructs a self-stabilizing extension channel around the balance constraint weight chain and forms a dynamic benefit response band in the benefit output stage. This enables the model to adaptively adjust the parameter influence ratio when facing a continuous input data stream, thereby allowing the benefit simulation process to maintain a gradual update rhythm in long-term operation. When sudden changes occur in the aquaculture environment, the system can achieve flexible buffering through the dynamic benefit response band, keeping the benefit prediction output structurally stable, ensuring a smooth and continuous prediction trend, and improving the self-stabilization of benefit simulation and the controllability of economic decision-making.

[0049] This invention provides, for example Figure 2 The shrimp farming revenue simulation system based on big data shown includes a multi-source monitoring data mapping module, an environmental variable mutation identification module, a balance constraint weight chain control module, a self-stabilizing extension channel construction module, and a dynamic revenue response band generation module. The multi-source monitoring data mapping module collects multi-source monitoring data from shrimp farming sites and performs continuous mapping processing on the multi-source monitoring data according to the time and spatial dimensions to construct a time-continuous feature mapping band. The environmental variable mutation identification module, based on the time-continuous feature mapping band, simultaneously calculates the rate of change of the corresponding environmental variable when new shrimp farming monitoring data enters the shrimp farming profit simulation process, and cuts the rate of change based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. The balance constraint weight chain regulation module introduces a balance constraint weight chain based on mutation identification fragments during the parameter update process of shrimp farming profit simulation. The changes in environmental variables before and after the mutation identification fragments are incorporated into the balance constraint weight chain for regulation. The self-stabilizing extended channel construction module builds a self-stabilizing extended channel around the balance constraint weight chain, and adjusts the influence ratio of mutation identification fragments and historical data in parameter updates in a linked manner. The dynamic profit response band generation module forms a dynamic profit response band based on a self-stabilizing extended channel. It introduces the parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extended channel into the shrimp farming profit simulation output process, and outputs the shrimp farming profit simulation results.

[0050] The big data-based shrimp farming profit simulation method provided in this embodiment of the invention is implemented through the above-mentioned big data-based shrimp farming profit simulation system. For details of the specific methods and processes of the big data-based shrimp farming profit simulation system, please refer to the above-mentioned embodiment of the big data-based shrimp farming profit simulation method, which will not be repeated here.

[0051] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for simulating shrimp farming profits based on big data, characterized in that, Includes the following steps: Step 1: Collect multi-source monitoring data from shrimp farming sites, and perform continuous mapping processing on the multi-source monitoring data according to the time and spatial dimensions to construct a time-continuous feature mapping band. Step 2: Based on the time-continuous feature mapping band, when new shrimp farming monitoring data enters the shrimp farming revenue simulation process, the rate of change of the corresponding environmental variable is calculated simultaneously, and the rate of change is cut based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. Step 3: In the parameter update process of shrimp farming profit simulation, a balance constraint weight chain is introduced based on the mutation identification fragment, and the changes in environmental variables before and after the mutation identification fragment are included in the balance constraint weight chain for regulation. Step 4: Construct a self-stabilizing extension channel around the balance constraint weight chain, and adjust the influence ratio of mutation identification fragments and historical data in parameter updates in a linked manner. Step 5: Based on the self-stabilizing extended channel, a dynamic profit response band is formed. The parameter update results after adjustment by the balance constraint weight chain and the self-stabilizing extended channel are introduced into the shrimp farming profit simulation output process to output the shrimp farming profit simulation results.

2. The method for simulating shrimp farming profits based on big data according to claim 1, characterized in that, The steps for constructing a time-continuous feature mapping band by continuously mapping multi-source monitoring data according to the time and spatial dimensions include: For various types of monitoring equipment deployed at shrimp farming sites, multi-source monitoring data is collected, and timestamps, spatial coordinates of collection points, equipment numbers, and collection depth information are added during the data recording stage. After data collection is completed, the multi-source monitoring data is continuously mapped according to the time dimension. The environmental variable change curves are formed by arranging the timestamps in order and interpolating to supplement the data. The time axes of different variables are then uniformly aligned. After obtaining the time-continuous characteristic curve, the spatial dimension of the multi-source monitoring data within the same time period is continuously mapped. The spatial change curve is established with the spatial coordinates of the aquaculture area as a reference, and the curves are superimposed to form a multivariate spatial distribution structure. After completing the continuous mapping of the time dimension and the continuous mapping of the spatial dimension, the continuous mapping of the time dimension and the continuous mapping of the spatial dimension are fused together, and the spatial distribution curves are stacked in the order of the time axis to construct the time continuous feature mapping band.

3. The method for simulating shrimp farming profits based on big data according to claim 1, characterized in that, The steps of segmenting the rate of change based on a preset rate threshold to form mutation identification fragments characterizing abrupt changes in environmental variables include: Based on the established time-continuous feature mapping band, when new shrimp farming monitoring data is input into the shrimp farming revenue simulation process, the new data is correlated in the time dimension, and its position in the time-continuous feature mapping band is determined according to the timestamp and time-series docked with adjacent historical data nodes. After completing the time series docking, the rate of change for each type of environmental variable is calculated, and the change per unit time is determined by combining the numerical difference between the new data point and the feature mapping data point of the previous moment with the time interval. After obtaining the rate of change of each environmental variable, a rate threshold is set according to the characteristics of the aquaculture environment and the physiological tolerance range of shrimp, and the rate results are cut off by the threshold to identify the environmental mutation interval. After identifying the environmental mutation interval, multiple variable data within the same time period are merged and recombined to form a mutation identification segment that includes time range, variable type, rate of change, and direction of change.

4. The method for simulating shrimp farming profits based on big data according to claim 3, characterized in that, The steps involved in incorporating the changes in environmental variables before and after the mutation identification fragment into the balance constraint weight chain for regulation include: Before the shrimp farming profit simulation enters the parameter update stage, time period matching and variable mapping are performed based on mutation identification segments. Stable time period data before the mutation and recovery time period data after the mutation are extracted to form a variable change data set containing three continuous data segments: before the mutation, during the mutation, and after the mutation. After obtaining environmental variable change data before and after the mutation identification segment, an initial structure of balanced constraint weight chain is established. With the mutation identification segment as the core, the magnitude and direction of change of each environmental variable before, during and after the mutation are incorporated into the weight relationship to form a continuous weight chain structure. After the balance constraint weight chain is established, the change in the variable corresponding to the mutation identification fragment is interactively controlled with the weight chain, and the parameter update rate and magnitude are gradually adjusted according to the weight ratio. After completing the balance constraint weight chain adjustment, the updated parameters after balance are mapped to the time-continuous feature map.

5. The method for simulating shrimp farming profits based on big data according to claim 4, characterized in that, When the balanced constraint weight chain interactively regulates the change in variables corresponding to the mutation identification segment, it adjusts the proportion of data in the three time periods before, during, and after the mutation in the weight chain. This allows the data in the pre-mutation time period to suppress parameter jumps, the data in the during-mutation time period to maintain a balanced transition, and the data in the post-mutation time period to guide the parameters back to a stable state.

6. The method for simulating shrimp farming profits based on big data according to claim 4, characterized in that, The steps involved in constructing a self-stabilizing extension channel around a balance constraint weight chain, and adjusting the influence ratio of mutation identification fragments and historical data in parameter updates through linkage, include: Under the premise that the balance constraint weight chain is formed and the mutation identification fragment is regulated, the mutation identification fragment and its corresponding historical data are classified and organized, the time interval is determined according to the weight node of the balance constraint weight chain, and the adjacent historical data sequences before and after the mutation are extracted as the balance reference interval. After forming a time-continuous input structure of mutation identification fragments and historical data, the influence relationship between the two types of data in parameter updates is preset in a linked manner, the influence range of mutation identification fragments and historical data is determined, and a dynamic weight allocation relationship is established. After establishing the linkage between mutation identification fragments and historical data, this linkage is introduced into the parameter update process, and the weight ratio between mutation identification fragments and historical data is automatically adjusted according to the time correlation of newly added shrimp farming monitoring data. After the linkage adjustment mechanism between mutation identification fragments and historical data is running stably, the self-stabilizing extension channel is continuously extended, and the influence ratio of mutation identification fragments and historical data is recorded in the time-continuous feature mapping band.

7. The method for simulating shrimp farming profits based on big data according to claim 6, characterized in that, After forming a time-continuous input structure of mutation identification fragments and historical data, when the influence relationship between the two types of data in parameter updates is preset in a linked manner, the influence range of the mutation identification fragments in the early stage of parameter updates covers the period of mutation occurrence and the short-term response interval, while the influence range of historical data covers the stable state before mutation and the recovery stage after mutation.

8. The method for simulating shrimp farming profits based on big data according to claim 6, characterized in that, The parameter update results, adjusted by the balance constraint weight chain and the self-stabilizing extended channel, are introduced into the shrimp farming profit simulation output process. The steps for outputting the shrimp farming profit simulation results include: Under the premise of stable operation of the self-stabilizing extended channel, the parameter update results after joint adjustment by the balance constraint weight chain and the self-stabilizing extended channel are integrated in the time dimension to form a continuous update sequence that reflects the dynamic change trend of parameters, and matched with the corresponding time continuous feature mapping band. After obtaining the adjusted parameter time-continuous structure, a spatial extension structure of the dynamic benefit response band is constructed based on the output path of the self-stabilizing extension channel. The spatial distribution characteristics of different parameters are synchronously mapped with the time-continuous parameter curve as the main axis, forming a smooth transition structure across regions. After establishing the characteristics of temporal continuity and spatial extension, the resulting dynamic profit response band is introduced into the shrimp farming profit simulation output process to synchronously adjust the variables in the profit calculation, so that the profit prediction forms a stable profit change curve under the drive of continuously changing parameters. Based on the dynamic return response band participating in the return output, the continuous results of return change trend are dynamically fed back and integrated. The output data is aligned and recorded with the time continuous feature mapping band, and used as a reference input in subsequent prediction periods.

9. The method for simulating shrimp farming profits based on big data according to claim 8, characterized in that, When introducing the shrimp farming revenue simulation output process into the dynamic revenue response band, the parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extension channel are used as the revenue calculation input according to the synchronous rhythm of the time continuous feature mapping band.

10. A big data-based shrimp farming profit simulation system, used to implement the big data-based shrimp farming profit simulation method according to any one of claims 1-9, characterized in that, It includes a multi-source monitoring data mapping module, an environmental variable mutation identification module, a balance constraint weight chain control module, a self-stabilizing extension channel construction module, and a dynamic benefit response band generation module; The multi-source monitoring data mapping module collects multi-source monitoring data from shrimp farming sites and performs continuous mapping processing on the multi-source monitoring data according to the time and spatial dimensions to construct a time-continuous feature mapping band. The environmental variable mutation identification module, based on the time-continuous feature mapping band, simultaneously calculates the rate of change of the corresponding environmental variable when new shrimp farming monitoring data enters the shrimp farming profit simulation process, and cuts the rate of change based on the preset rate threshold to form a mutation identification segment that characterizes the mutation interval of the environmental variable. The balance constraint weight chain regulation module introduces a balance constraint weight chain based on mutation identification fragments during the parameter update process of shrimp farming profit simulation. The changes in environmental variables before and after the mutation identification fragments are incorporated into the balance constraint weight chain for regulation. The self-stabilizing extended channel construction module constructs a self-stabilizing extended channel around the balanced constraint weight chain, and adjusts the influence ratio of mutation identification fragments and historical data in parameter updates in a linked manner. The dynamic profit response band generation module forms a dynamic profit response band based on a self-stabilizing extended channel. It introduces the parameter update results after adjustment by the balanced constraint weight chain and the self-stabilizing extended channel into the shrimp farming profit simulation output process, and outputs the shrimp farming profit simulation results.

Images