Fish pond water temperature prediction method based on smart grid prediction data

Abstract

The application discloses a fishpond water temperature prediction method based on intelligent grid prediction data, and comprises the following steps: constructing a fishpond grid map, acquiring M candidate site coordinates in the same coordinate system as the fishpond grid map, determining the previous day's real-time air temperature and the current day's predicted air temperature of a target grid according to the M candidate site coordinates, selecting K anchor grids in the fishpond grid map, determining the previous day's real-time water temperature of the target grid according to the K anchor grids, determining the current day's predicted water temperature of the target grid according to the previous day's real-time water temperature, the previous day's real-time air temperature and the current day's predicted air temperature of the target grid, inputting the current day's predicted water temperature of the target grid into a drawing tool to obtain a water temperature prediction grid distribution map of the fishpond; the application can comprehensively cover the water temperature change in the fishpond, avoids the error caused by the excessively large grid or the local difference being ignored in the traditional prediction method, and provides more accurate water temperature prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to a water temperature prediction method, in particular to a fish pond water temperature prediction method based on intelligent grid prediction data. BACKGROUND

[0002] Currently, many aquaculture farmers need to actually predict the water temperature of the day when carrying out aquaculture. The patent document with the patent publication number CN202411682670.6 discloses a freshwater fish pond water temperature prediction method, system, medium and equipment, which can construct freshwater fish pond water temperature prediction models such as low-temperature oligo-illumination type, sharp cooling and pressure reduction type, and high-temperature hot type under different weather types. However, in the above patent document and prior art, water temperature prediction is usually based on point-based measured data. With the expansion and increasing complexity of water areas, traditional methods are difficult to provide sufficient spatial resolution to accurately predict the water temperature distribution of the entire water body. Especially in irregularly shaped water areas such as fish ponds, traditional air temperature data and water temperature models are difficult to accurately capture the temperature differences in each area, resulting in limited accuracy and practicality of water temperature prediction results.

[0003] In addition, existing water temperature prediction techniques rely on relatively rough air temperature and water temperature data, and usually cannot effectively consider factors such as water flow direction, water body terrain slope, and internal flow pattern. Therefore, the existing technology is often difficult to provide accurate water temperature distribution map in complex environment, and cannot fully reflect the water temperature difference between different grid points, thereby affecting the management of aquaculture. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a fish pond water temperature prediction method based on intelligent grid prediction data, which introduces grid water temperature prediction to solve the technical problems raised in the background art.

[0005] To achieve the above purpose, the present application is realized by the following technical scheme:

[0006] The fish pond water temperature prediction method based on intelligent grid prediction data comprises the following steps:

[0007] S1, constructing a fish pond grid map;

[0008] Among them, the fish pond grid map includes a plurality of grid coordinates;

[0009] S2, obtaining M candidate site coordinates in the same coordinate system as the fish pond grid map;

[0010] S3, determining the previous day's real-time air temperature and the current day's predicted air temperature of the target grid according to the M candidate site coordinates;

[0011] S4, selecting K anchor grids in the fish pond grid map;

[0012] S5, determining the previous day live water temperature of the target grid according to the K anchor grids;

[0013] S6, determining the current day forecast water temperature of the target grid according to the previous day live water temperature of the target grid, the previous day live air temperature and the current day forecast air temperature;

[0014] S7, inputting the current day forecast water temperature of the target grid into a drawing tool to obtain a water temperature forecast grid distribution map of the fish pond.

[0015] In some specific embodiments, the fish pond grid map is constructed, comprising:

[0016] S1-1, obtaining the boundary point coordinates of the fish pond;

[0017] S1-2, determining the fish pond center coordinates according to the boundary point coordinates of the fish pond;

[0018] S1-3, taking the fish pond center coordinates as the drawing origin point to construct the origin fish pond grid with a predefined grid side length; wherein the diagonal intersection point of the origin fish pond grid coincides with the drawing origin point;

[0019] S1-4, repeatedly constructing the fish pond grid on the four neighborhoods of the origin fish pond grid until the fish pond grid distribution covers at least all the boundary point coordinates;

[0020] S1-5, defining the fish pond grid distribution covering all the boundary point coordinates as the fish pond grid map.

[0021] In some specific embodiments, the fish pond grid is repeatedly constructed on the four neighborhoods of the origin fish pond grid until the fish pond grid distribution covers at least all the boundary point coordinates; comprising:

[0022] S1-4-1, obtaining the fish pond center coordinates and the four equal side lengths of the origin fish pond grid;

[0023] S1-4-2, marking the side length midpoint coordinates for any equal side length;

[0024] S1-4-3, extending the neighborhood virtual line from the fish pond center coordinates to the direction of the side length midpoint coordinates;

[0025] S1-4-4, calculating the extension length of the neighborhood virtual line;

[0026] S1-4-5, if the extension length of the neighborhood virtual line is twice the distance from the fish pond center coordinates, then stop extending the neighborhood virtual line, and define the end of the neighborhood virtual line as the neighborhood grid coordinates;

[0027] S1-4-6, constructing the neighborhood grid according to the neighborhood grid coordinates and the predefined grid side length;

[0028] S1-4-7, starting from the neighborhood grid coordinates, extending the neighborhood dotted line in the direction where the neighborhood grid is not constructed, repeating S1-4-4 to S1-4-7 until all boundary point coordinates are covered by the neighborhood grid.

[0029] In some specific embodiments, the M candidate site coordinates in the same coordinate system as the fish pond grid map are obtained, including:

[0030] S2-1, obtaining N air temperature measurement site coordinates in the same coordinate system as the fish pond grid map;

[0031] S2-2, calculating N first coordinate distances between N air temperature measurement site coordinates and the fish pond center coordinates;

[0032] S2-3, selecting M second coordinate distances from N first coordinate distances;

[0033] S2-4, obtaining M candidate site coordinates corresponding to the second coordinate distance from N air temperature measurement site coordinates.

[0034] In some specific embodiments, selecting M second coordinate distances from N first coordinate distances includes:

[0035] S2-3-1, sorting N first coordinate distances in ascending order to generate a first sequence;

[0036] S2-3-2, selecting the first coordinate distance with the shortest distance from the first sequence as the second coordinate distance, and adding it to the second sequence;

[0037] S2-3-3, repeating S2-3-2 until there are M second coordinate distances in the second sequence.

[0038] In some specific embodiments, according to the M candidate site coordinates, the previous day live air temperature and the current day forecast air temperature of the target grid are determined, including:

[0039] S3-1, obtaining the grid coordinates of each fish pond grid in the fish pond grid map;

[0040] S3-2, calculating the Euclidean distance between the target grid and the M candidate site coordinates;

[0041] S3-3, obtaining the previous day live air temperature and the current day forecast air temperature of the M candidate site coordinates;

[0042] S3-4, according to the previous day live air temperature and the current day forecast air temperature of the M candidate site coordinates, and the Euclidean distance between the target grid and the M candidate site coordinates, determining the previous day live air temperature and the current day forecast air temperature of the target grid;

[0043] The calculation formula of the previous day live air temperature of the target grid is:

[0044] ;

[0045] The calculation formula of the current day forecast air temperature of the target grid is:

[0046] ;

[0047] wherein, represents the previous day live air temperature of the grid coordinate represents the current day forecast air temperature of the grid coordinate represents the previous day live air temperature of the i-th candidate site, represents the current day forecast air temperature of the i-th candidate site, represents the Euclidean distance between the grid coordinate and the i-th candidate site, represents the distance decay index, which means that the farther the distance of the candidate site, the smaller the site weight.

[0048] In some specific embodiments, K anchor grids are selected in the fishpond grid map, including:

[0049] S4-1, calculating the Euclidean distance between the target grid and all boundary point coordinates;

[0050] S4-2, determining the average distance between the target grid and all boundary point coordinates according to the Euclidean distance between the target grid and all boundary point coordinates;

[0051] S4-3, calculating the standard deviation of the target grid according to the average distance and the Euclidean distance between the target grid and all boundary point coordinates, until a standard deviation set of all grid coordinates is obtained;

[0052] S4-4, selecting the maximum standard deviation from the standard deviation set of all grid coordinates in turn, and matching the corresponding grid as an anchor grid based on the maximum standard deviation;

[0053] S4-5, repeating 4-4 until K anchor grids are obtained.

[0054] In some specific embodiments, the previous day live water temperature of the target grid is determined according to the K anchor grids, including:

[0055] S5-1, calculating the Euclidean distance between the target grid and the K anchor grids;

[0056] S5-2, obtaining the previous day measured water temperature of the K anchor grids;

[0057] ​​S5-3, determining the previous day live water temperature of the target grid according to the previous day measured water temperature of the K anchor grids and the Euclidean distance between the target grid and the K anchor grids;

[0058] The calculation formula for determining the previous day live water temperature of the target grid is:

[0059] ;

[0060] Wherein, represents the previous day live water temperature of the target grid, represents the monitored previous day measured water temperature of the i th anchor grid, represents the Euclidean distance between the target grid and the anchor grid, represents the flow direction angle between the i th anchor grid and the target grid; represents the flow direction angle of the target grid, represents the water bottom slope difference between the target grid and the i th anchor grid, and k represents the influence coefficient of controlling the water bottom slope difference value

[0061] The present application provides a fish pond water temperature prediction method based on intelligent grid prediction data, which has the following beneficial effects:

[0062] The present application significantly improves the spatial accuracy of water temperature prediction by constructing a fish pond grid map and combining data from multiple candidate sites. Fish ponds are usually irregular in shape, and traditional methods use relatively simple gridding processing, which is difficult to effectively capture the changes of water temperature in different areas. Therefore, by dividing the fish pond into multiple grid units, the water temperature of each grid unit can be predicted separately, ensuring that the temperature change in each area can be predicted. This method can fully cover the water temperature changes in the fish pond, avoiding the errors caused by large grid size or ignoring local differences in traditional prediction methods, and providing more accurate water temperature prediction.

[0063] Further, by introducing the flow direction angle and the water bottom slope difference in the process of predicting the water temperature of the target grid, the calculation of the water temperature is optimized, and the prediction accuracy of the live water temperature is further improved. By introducing the flow direction angle, the propagation and variation of water temperature in the water body can be simulated, ensuring that the water temperature prediction is more consistent with the actual flow characteristics of the water body. The introduction of the water bottom slope difference considers the influence of the terrain on the water temperature change, and the water temperature change will be more intense in areas with larger slope difference. Especially in the case of irregular water body shape or complex water flow environment, the water temperature prediction result can be dynamically adjusted to reflect the overall picture of the water temperature change in the water body in real time. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 is a flowchart of the fish pond water temperature prediction method based on intelligent grid prediction data of the present application;

[0065] Figure 2 This is a schematic diagram illustrating the construction process of the fishpond grid map described in this invention;

[0066] Figure 3 This is a schematic diagram of the selection process for the anchoring grid described in this invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Example 1: See Figure 1 This invention provides a method for predicting fishpond water temperature based on intelligent grid forecast data, comprising the following steps:

[0069] S1. Construct a grid map of the fishpond;

[0070] The fishpond grid map includes several grid coordinates;

[0071] S2. Obtain the coordinates of M candidate stations located in the same coordinate system as the fishpond grid map;

[0072] S3. Based on the coordinates of M candidate stations, determine the previous day's actual temperature and the current day's forecast temperature for the target grid;

[0073] It should be noted that the daily forecast temperature refers to the expected temperature value calculated based on existing temperature prediction models. This expected temperature value takes into account the weather conditions at that time, historical data, and other factors that affect temperature changes. Therefore, the daily forecast temperature represents the expected temperature of the area where the fishpond is located during a specific period.

[0074] S4. Select K anchor grids in the fishpond grid map;

[0075] S5. Based on the K anchored grids, determine the actual water temperature of the target grid the previous day;

[0076] S6. Based on the previous day's actual water temperature, the previous day's actual air temperature, and the current day's forecast air temperature, determine the current day's forecast water temperature for the target grid.

[0077] For example, in this embodiment, the formula for calculating the daily forecast water temperature of the target grid is:

[0078] ;

[0079] in, This indicates the predicted water temperature for the day. This indicates the forecast temperature for the day. This indicates the actual temperature of the previous day. This indicates the actual water temperature of the previous day; This is a constant term used to represent the reference offset. , , The weights of the influence of the forecast temperature of the day, the actual temperature of the previous day, and the actual water temperature of the previous day on the forecast water temperature of the day are respectively.

[0080] For example, constant term The constant term is fixed in the prediction model but can fluctuate within a reasonable range. In this embodiment, the given constant term is 3.111, and this value typically fluctuates within ±0.5. Therefore, the range of the constant term can be defined as [2.611, 3.611]. Specifically, the constant term... It is an offset derived from historical data, used to correct the baseline relationship between air and water temperatures. Its upper and lower fluctuation range is set based on the fluctuation range of actual meteorological data.

[0081] Furthermore, the influence weight , , These represent the relative importance of the forecast air temperature, the previous day's actual air temperature, and the previous day's actual water temperature in the calculation of the forecast water temperature for the current day. For example, in this embodiment, , , .

[0082] The specific numerical values ​​affecting the weights are explained in this embodiment as follows:

[0083] The forecasted air temperature has a relatively small impact on water temperature, with a coefficient of 0.214. This is because while air temperature does have some influence on water temperature, changes in water temperature are not solely caused by the current day's air temperature but are also affected by the previous day's air and water temperatures. Furthermore, water has greater thermal inertia, meaning its temperature does not change as rapidly as air temperature. Therefore, A value of 0.214 indicates that the predicted air temperature contributes little to the water temperature.

[0084] The previous day's actual air temperature has a significant impact on water temperature, with a coefficient of 0.523. Water temperature is a physical quantity with strong thermal inertia, and the previous day's air temperature usually has a significant influence on the current water temperature. This is because water temperature changes take time, and the previous day's air temperature has a significant cumulative effect on water temperature, especially in the absence of drastic weather changes; the previous day's air temperature determines the baseline water temperature level. Therefore, The value was set to 0.523, indicating that the previous day's actual air temperature had the strongest impact on the forecast water temperature for the current day.

[0085] The previous day's actual water temperature has the least impact, with a coefficient of 0.147. The influence of the previous day's water temperature on water temperature is continuous, but relatively weak, because water temperature adjusts with changes in air temperature and the environment. The previous day's water temperature provides some initial conditions for the current water temperature, but because the water body has time to adapt to changes in air temperature, the dependence of water temperature on the previous day's temperature is relatively small. Therefore, Set as This indicates that the previous day's water temperature has a relatively weak impact on the predicted water temperature.

[0086] S7. Input the daily forecast water temperature of the target grid into the output tool to obtain the water temperature forecast grid distribution map of the fishpond.

[0087] Specifically, in this embodiment, the water temperature forecast grid distribution map refers to a grid distribution map formulated based on the daily forecast air temperature of the target grid. Specifically, this grid distribution map represents the water temperature forecast value for each grid in the entire fishpond. Each grid in the map represents a specific spatial unit, and the water temperature value of each grid is calculated using a model based on its surrounding environmental conditions (such as air temperature, flow direction, slope, etc.).

[0088] This embodiment constructs a fishpond grid map and combines data from multiple candidate stations to predict the water temperature of each grid within the fishpond. The method utilizes the spatial relationship between the target grid and multiple candidate stations, combined with factors such as the previous day's actual air temperature, the current day's forecast air temperature, and the previous day's actual water temperature, to calculate the current day's forecast water temperature for each grid. Ultimately, the generated water temperature forecast grid distribution map visually displays the water temperature distribution in different areas of the fishpond, providing data support for fishpond aquaculture management.

[0089] Example 2: See Figures 2 to 3 The technical solution of Embodiment 2 differs from Embodiment 1 in that it discloses the following exemplary steps. Specifically, the construction of the fishpond grid map includes:

[0090] S1-1. Obtain the coordinates of the boundary points of the fishpond;

[0091] S1-2. Determine the coordinates of the center of the fishpond based on the coordinates of the boundary points of the fishpond;

[0092] It should be noted that fishponds are mostly irregular planar shapes; therefore, the center coordinates of the fishpond can be calculated based on the coordinates of the boundary points. Specifically, if the boundary of the fishpond is an irregular polygon (such as an uneven shape), the centroid method can be used to calculate the center coordinates. The centroid is the weighted average position of all points in the polygon. The centroid method is suitable for irregularly shaped water areas because it can determine the geometric center of the fishpond by considering the distribution of all boundary points, rather than just a simple geometric center point. In practice, the boundary point coordinates of the fishpond may not be completely closed; it is necessary to first compare the boundary point coordinates to ensure the closure operation and thus the accuracy of the calculation.

[0093] S1-3. Using the center coordinates of the fishpond as the origin, construct the origin fishpond grid with a predefined grid side length; wherein, the intersection point of the diagonals of the origin fishpond grid coincides with the origin.

[0094] S1-4. On the four neighboring regions of the origin fishpond grid, repeat the construction of the fishpond grid until the fishpond grid distribution covers at least all boundary point coordinates.

[0095] S1-5. Define the fishpond grid distribution that covers the coordinates of all boundary points as a fishpond grid map.

[0096] Since fishponds are typically irregularly shaped, this embodiment employs the centroid method to determine the center coordinates of the fishpond. The centroid method accurately calculates the geometric center of the fishpond based on the coordinates of its boundary points. By gridding the area around the center coordinates of the fishpond according to predefined grid side lengths and gradually expanding to cover the entire fishpond area, a complete fishpond grid map is ultimately constructed. Through this gridding method, each grid cell in the spatial distribution of the fishpond acts as an independent computational unit, effectively capturing water temperature changes in different grid areas.

[0097] For example, the step of repeatedly constructing a fishpond grid on the four neighboring regions of the origin fishpond grid until the fishpond grid distribution at least covers all boundary point coordinates includes:

[0098] S1-4-1. Obtain the coordinates of the center of the fishpond, and the four equal side lengths of the fishpond grid at the origin;

[0099] S1-4-2. For any side of equal length, mark the coordinates of its midpoint;

[0100] S1-4-3. Starting from the center coordinates of the fishpond, extend the neighborhood dashed line in the direction of the midpoint coordinates of the side length;

[0101] S1-4-4, Calculate the extension length of the neighboring dashed line;

[0102] S1-4-5. If the extension length of the neighboring dashed line is twice the coordinates of the center of the fishpond, then stop the extension of the neighboring dashed line and define the end of the neighboring dashed line as the coordinates of the neighboring grid.

[0103] S1-4-6. Construct a neighborhood grid based on the neighborhood grid coordinates and the predefined grid edge length;

[0104] S1-4-7. Starting from the coordinates of the neighboring grid, extend the neighboring dashed line in the direction where no neighboring grid has been built. Repeat S1-4-4 to S1-4-7 until all boundary point coordinates are covered by the neighboring grid.

[0105] This embodiment ensures that the fishpond grid distribution covers the entire fishpond boundary by progressively expanding the grid in the four neighborhoods of the origin fishpond grid. Specifically, the center coordinates of the fishpond and the four equal side lengths of the origin fishpond grid are first determined. By marking the midpoint coordinates of each side length, a neighborhood dashed line extending from the center of the fishpond in each side length direction is established, and the extension length is calculated. The dashed line stops extending to twice the distance from the center of the fishpond, and the end of the dashed line is defined as the neighborhood grid coordinates. Finally, the complete fishpond grid coverage is progressively expanded, ensuring that all boundary points are included. By accurately covering all boundary points and progressively expanding the grid, the problem of incomplete grid coverage caused by irregular shapes is effectively avoided.

[0106] For example, obtaining the coordinates of M candidate stations located in the same coordinate system as the fishpond grid map includes:

[0107] S2-1. Obtain the coordinates of N temperature measurement stations located in the same coordinate system as the fishpond grid map;

[0108] S2-2. Calculate the distances between the coordinates of N temperature measurement stations and the N first coordinates of the fishpond center.

[0109] S2-3. Select M second coordinate distances from N first coordinate distances;

[0110] S2-4. From the coordinates of N temperature measurement stations, obtain the coordinates of M candidate stations corresponding to the second coordinate distance.

[0111] Specifically, the spatial relationship between the fishpond center and all stations is obtained by calculating the first coordinate distance between each temperature measurement station and the center of the fishpond. Then, based on the calculated first coordinate distance, M stations with the shortest distance to the fishpond center are selected. These M stations are considered to have the greatest impact on changes in fishpond water temperature. Selecting closer stations as candidate stations ensures the correlation between the selected data and fishpond water temperature.

[0112] For example, selecting M second coordinate distances from N first coordinate distances includes:

[0113] S2-3-1. Sort the N first coordinate distances in ascending order to generate the first sequence;

[0114] S2-3-2. Select the shortest first coordinate distance from the first sequence as the second coordinate distance and add it to the second sequence;

[0115] S2-3-3, Repeat S2-3-2 until there are M second coordinate distances in the second sequence.

[0116] In this embodiment, a first sequence is generated by arranging N first coordinate distances in ascending order. Then, the coordinate with the shortest distance from the first sequence is selected as the second coordinate distance, and this is added to the second sequence sequentially until the second sequence contains M shortest coordinate distances. This step ensures that the M stations with the shortest distance to the center of the fishpond are selected from the candidate stations to maximize the correlation with water temperature changes.

[0117] For example, determining the previous day's actual temperature and the current day's forecast temperature of the target grid based on the coordinates of M candidate stations includes:

[0118] S3-1. Obtain the grid coordinates of each fishpond grid in the fishpond grid map;

[0119] S3-2. Calculate the Euclidean distance between the target grid and the coordinates of the M candidate sites;

[0120] S3-3. Obtain the previous day's actual temperature and the current day's forecast temperature for the coordinates of M candidate stations;

[0121] S3-4. Based on the previous day's actual temperature and the current day's forecast temperature of the M candidate station coordinates, and the Euclidean distance between the target grid and the M candidate station coordinates, determine the previous day's actual temperature and the current day's forecast temperature of the target grid.

[0122] The formula for calculating the previous day's actual temperature of the target grid is:

[0123] ;

[0124] The formula for calculating the daily forecast temperature of the target grid is:

[0125] ;

[0126] in, Represents grid coordinates The previous day's real temperature Represents grid coordinates The forecast temperature for the day, This represents the actual temperature of the i-th candidate station on the previous day. This represents the daily forecast temperature for the i-th candidate station. Represents grid coordinates The Euclidean distance between the i-th candidate site and the i-th candidate site is... , and The coordinates of the first candidate site; The distance decay index is 2 in this embodiment, which means that the farther away the candidate site is, the smaller its site weight.

[0127] In other words, for each grid station corresponding to a fishpond grid, both the previous day's actual temperature and the current day's forecast temperature are determined by the Euclidean distance between the station and the M candidate stations. This distance... The inverse weighting means that the closer the candidate station is, the greater its influence on the temperature of the grid point, thus ensuring that the temperature of each grid point is determined by weighted interpolation from multiple stations.

[0128] In this embodiment, the previous day's actual temperature and the current day's forecast temperature for the target grid are calculated using distance-weighted interpolation. Specifically, the coordinates of each fishpond grid are first obtained, and the Euclidean distance between the grid and M candidate stations is calculated. Then, the previous day's actual temperature and the current day's forecast temperature for each candidate station are obtained, and the data of each candidate station are weighted according to the inverse weighting method based on the Euclidean distance to ensure that stations with closer distances have a greater impact on the temperature of the target grid, thereby improving the relevance of the temperature calculation.

[0129] For example, selecting K anchor grids in the fishpond grid map includes:

[0130] S4-1. Calculate the Euclidean distance between the target mesh and the coordinates of all boundary points;

[0131] S4-2. Determine the average distance between the target grid and all boundary point coordinates based on the Euclidean distance between the target grid and all boundary point coordinates;

[0132] S4-3. Calculate the standard deviation of the target grid based on the average distance and Euclidean distance between the target grid and the coordinates of all boundary points, until the set of standard deviations of all grid coordinates is obtained;

[0133] S4-4. Select the largest standard deviation from the set of standard deviations of all grid coordinates in sequence, and match the corresponding grid as the anchor grid based on the largest standard deviation;

[0134] S4-5, Repeat 4-4 until K anchored grids are obtained.

[0135] In this embodiment, the standard deviation is introduced to evaluate the spatial uniformity of the target grid and boundary point coordinates. Specifically, the Euclidean distances between the target grid and all boundary point coordinates are calculated, and the standard deviation of each grid is calculated based on these distances. The standard deviation reflects the dispersion of the distances between the target grid and the boundary point coordinates. A grid with a larger standard deviation indicates a greater difference in distance from the second boundary point, meaning that the grid is spatially more uniformly distributed, neither too close to nor too far from the boundary. The selection of the maximum standard deviation ensures that the selected anchor grid has a more uniform spatial distribution because it does not deviate too far from the boundary while not being too close to it, thus it can be defined as a grid representing the temperature changes in the fishpond area. Through this step, the spatial distribution of the anchor grid is more balanced, avoiding prediction errors caused by anchor grids that are too concentrated or deviated from the boundary points.

[0136] For example, determining the previous day's actual water temperature of the target grid based on K anchored grids includes:

[0137] S5-1. Calculate the Euclidean distance between the target mesh and the K anchored meshes;

[0138] S5-2. Obtain the measured water temperature of the K anchoring grids the day before;

[0139] It should be noted that the water temperature measured the day before the anchoring grid is real-time data collected by water temperature sensors or temperature detection instruments within the fishpond. These devices are deployed at different locations within the fishpond to collect actual water temperature information. The water temperature data from the previous day reflects the actual measured temperature conditions.

[0140] S5-3. Based on the measured water temperature of the K anchored grids the previous day and the Euclidean distance between the target grid and the K anchored grids, determine the actual water temperature of the target grid the previous day.

[0141] The formula for calculating the actual water temperature of the day before determining the target grid is:

[0142] ;

[0143] in, This indicates the actual water temperature of the target grid the day before (i.e., the actual water temperature of the target grid the day before). This represents the measured water temperature of the i-th anchored grid from the previous day. This represents the Euclidean distance between the target mesh and the anchor mesh. This represents the angle of the water flow direction between the i-th anchored grid and the target grid; Indicates the angle of the water flow direction in the target grid. This represents the difference in bottom slope between the target grid and the i-th anchored grid, where k represents the influence coefficient controlling the difference in bottom slope.

[0144] Specifically, this calculation formula incorporates the correlation of the difference in the angle of the water flow direction between the target grid and the anchored grid. If the water flow directions are the same, the correlation is high; conversely, if the water flow directions differ significantly, the correlation is low. Furthermore, it also incorporates the correlation of the difference in bottom slope. That is, the greater the difference in bottom slope between the target grid and the anchored grid, the more drastic the water temperature change. In other words, the farther the anchored grid is, the less influence it has on the water temperature of the target grid; the closer the anchored grid is to the water flow direction, the greater its influence on the water temperature of the target grid; and the greater the difference in bottom slope, the greater the influence of the anchored grid on the water temperature of the target grid.

[0145] In this embodiment, the previous day's actual water temperature of the target grid is calculated by extrapolating the previous day's measured water temperature of the anchor grid and related environmental factors. Although the previous day's actual water temperature of the anchor grid is collected in real time at multiple locations within the fishpond using water temperature sensors or temperature detection instruments, the number of grids in the fishpond is usually extremely large, making it impractical to measure the water temperature of each grid. In other words, the number of target grids is typically much larger than the number of anchor grids. Therefore, it is necessary to use the previous day's actual temperature of the anchor grid to extrapolate the water temperature of other unmeasured areas.

[0146] In this embodiment, Euclidean distance is used to measure the spatial proximity between the target grid and the anchor grid; the closer the grid is, the greater its influence on the water temperature of the target grid. Secondly, the water flow direction angle considers the influence of water flow on water temperature transfer; the closer the water flow direction, the stronger the influence of the grid on the water temperature of the target grid. Finally, the difference in bottom slope reflects the moderating effect of topography on water temperature changes; areas with greater slope differences experience more drastic water temperature changes, and their weight increases accordingly. By introducing these parameters, this embodiment can more accurately calculate the previous day's actual water temperature of the target grid based on the previous day's measured water temperature of the anchor grid, thereby ensuring higher prediction accuracy and greater spatial consistency of the previous day's actual water temperature of the target grid.

[0147] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means.

[0148] The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g.,...), etc. DVD ( ), or semiconductor media. Semiconductor media can be solid-state drives (SSDs).

[0149] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Furthermore, the mutual couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0150] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for predicting fishpond water temperature based on intelligent grid forecast data, characterized in that, include: S1. Construct a grid map of the fishpond; The fishpond grid map includes several grid coordinates; S2. Obtain the coordinates of M candidate stations located in the same coordinate system as the fishpond grid map; S3. Based on the coordinates of M candidate stations, determine the previous day's actual temperature and the current day's forecast temperature for the target grid; S4. Select K anchor grids in the fishpond grid map; The step of selecting K anchor grids in the fishpond grid map includes: S4-1. Calculate the Euclidean distance between the target mesh and the coordinates of all boundary points; S4-2. Determine the average distance between the target grid and all boundary point coordinates based on the Euclidean distance between the target grid and all boundary point coordinates; S4-3. Calculate the standard deviation of the target grid based on the average distance and Euclidean distance between the target grid and the coordinates of all boundary points, until the set of standard deviations of all grid coordinates is obtained; S4-4. Select the largest standard deviation from the set of standard deviations of all grid coordinates in sequence, and match the corresponding grid as the anchor grid based on the largest standard deviation; S4-5, Repeat S4-4 until K anchoring grids are obtained; The anchoring grid is defined as a grid representing the temperature changes in the fishpond area; S5. Based on the K anchored grids, determine the actual water temperature of the target grid the previous day; The step of determining the previous day's actual water temperature of the target grid based on K anchored grids includes: S5-1. Calculate the Euclidean distance between the target mesh and the K anchored meshes; S5-2. Obtain the measured water temperature of the K anchoring grids the day before; S5-3. Based on the measured water temperature of the K anchored grids the previous day and the Euclidean distance between the target grid and the K anchored grids, determine the actual water temperature of the target grid the previous day. The formula for calculating the actual water temperature of the day before determining the target grid is: ； in, This indicates the actual water temperature of the target grid the previous day. This represents the measured water temperature of the i-th anchored grid from the previous day. This represents the Euclidean distance between the target mesh and the anchor mesh. This represents the angle of the water flow direction between the i-th anchored grid and the target grid; Indicates the angle of the water flow direction in the target grid. This represents the difference in seabed slope between the target grid and the i-th anchored grid, where k represents the influence coefficient controlling the value of the seabed slope difference. Indicates the distance decay index; S6. Based on the previous day's actual water temperature, the previous day's actual air temperature, and the current day's forecast air temperature, determine the current day's forecast water temperature for the target grid. S7. Input the daily forecast water temperature of the target grid into the output tool to obtain the water temperature forecast grid distribution map of the fishpond.

2. The method for predicting fishpond water temperature based on intelligent grid forecast data according to claim 1, characterized in that, Constructing a fishpond grid map includes: S1-1. Obtain the coordinates of the boundary points of the fishpond; S1-2. Determine the coordinates of the center of the fishpond based on the coordinates of the boundary points of the fishpond; S1-3. Using the center coordinates of the fishpond as the origin, construct the origin fishpond grid with a predefined grid side length; wherein, the intersection point of the diagonals of the origin fishpond grid coincides with the origin. S1-4. On the four neighboring regions of the origin fishpond grid, repeat the construction of the fishpond grid until the fishpond grid distribution covers at least all boundary point coordinates. S1-5. Define the fishpond grid distribution that covers the coordinates of all boundary points as a fishpond grid map.

3. The method for predicting fishpond water temperature based on intelligent grid forecast data according to claim 2, characterized in that, On the four neighboring regions of the origin's fishpond mesh, repeat the fishpond mesh construction until the fishpond mesh distribution at least covers the coordinates of all boundary points; including: S1-4-1. Obtain the coordinates of the center of the fishpond, and the four equal side lengths of the fishpond grid at the origin; S1-4-2. For any side of equal length, mark the coordinates of its midpoint; S1-4-3. Starting from the center coordinates of the fishpond, extend the neighborhood dashed line in the direction of the midpoint coordinates of the side length; S1-4-4, Calculate the extension length of the neighboring dashed line; S1-4-5. If the extension length of the neighboring dashed line is twice the coordinates of the center of the fishpond, then stop the extension of the neighboring dashed line and define the end of the neighboring dashed line as the coordinates of the neighboring grid. S1-4-6. Construct a neighborhood grid based on the neighborhood grid coordinates and the predefined grid edge length; S1-4-7. Starting from the coordinates of the neighboring grid, extend the neighboring dashed line in the direction where no neighboring grid has been built. Repeat S1-4-4 to S1-4-7 until all boundary point coordinates are covered by the neighboring grid.

4. The method for predicting fishpond water temperature based on intelligent grid forecast data according to claim 1, characterized in that, Obtain the coordinates of M candidate stations located in the same coordinate system as the fishpond grid map, including: S2-1. Obtain the coordinates of N temperature measurement stations located in the same coordinate system as the fishpond grid map; S2-2. Calculate the distances between the coordinates of N temperature measurement stations and the N first coordinates of the fishpond center. S2-3. Select M second coordinate distances from N first coordinate distances; S2-4. From the coordinates of N temperature measurement stations, obtain the coordinates of M candidate stations corresponding to the second coordinate distance.

5. The method for predicting fishpond water temperature based on intelligent grid forecast data according to claim 1, characterized in that, Choose M second coordinate distances from N first coordinate distances, including: S2-3-1. Sort the N first coordinate distances in ascending order to generate the first sequence; S2-3-2. Select the shortest first coordinate distance from the first sequence as the second coordinate distance and add it to the second sequence; S2-3-3, Repeat S2-3-2 until there are M second coordinate distances in the second sequence.

6. The method for predicting fishpond water temperature based on intelligent grid forecast data according to claim 1, characterized in that, Based on the coordinates of M candidate stations, determine the previous day's actual temperature and the current day's forecast temperature for the target grid, including: S3-1. Obtain the grid coordinates of each fishpond grid in the fishpond grid map; S3-2. Calculate the Euclidean distance between the target grid and the coordinates of the M candidate sites; S3-3. Obtain the previous day's actual temperature and the current day's forecast temperature for the coordinates of M candidate stations; S3-4. Based on the previous day's actual temperature and the current day's forecast temperature of the M candidate station coordinates, and the Euclidean distance between the target grid and the M candidate station coordinates, determine the previous day's actual temperature and the current day's forecast temperature of the target grid. The formula for calculating the previous day's actual temperature of the target grid is: ； The formula for calculating the daily forecast temperature of the target grid is: ； in, Represents grid coordinates The previous day's real temperature Represents grid coordinates The forecast temperature for the day, This represents the actual temperature of the i-th candidate station on the previous day. This represents the daily forecast temperature for the i-th candidate station. Represents grid coordinates The Euclidean distance between the i-th candidate site and the i-th candidate site. This represents the distance decay index, indicating that the farther away a candidate site is, the lower its site weight.

Images