Marine area environment temperature monitoring method and system based on coastal culture area

Abstract

The invention provides an ocean area environment temperature monitoring method and system based on a coastal culture area, and relates to the technical field of data processing, and the method comprises the steps: carrying out the multi-dimensional element analysis of fused data, so as to obtain an analysis result; according to an analysis result, performing space-time association and attribute association analysis on the parameter data of the marine environment to obtain marine space-time information; performing fuzzy division on the marine spatio-temporal information to obtain dynamic change feature data of the marine environment; the method comprises the following steps: establishing spatial distribution of regular grid average scatter dynamic change feature data to obtain spatial gradient distribution features of ocean data; according to the spatial gradient distribution characteristics, the change trend of the marine environment temperature is recognized, and environment early warning is provided for the coastal culture area. According to the invention, timely and accurate environment information can be provided for coastal culture areas.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for monitoring the environmental temperature of a marine area based on a coastal aquaculture area. Background Art

[0002] As an important part of the marine economy, the stable development of coastal aquaculture areas is of great significance to the sustainable use of marine resources. However, due to the complexity and variability of the marine environment, especially the fluctuation of marine environmental temperature, coastal aquaculture activities have brought many uncertainties. This uncertainty not only affects the growth cycle and yield of aquaculture organisms, but may also cause a series of ecological and environmental problems, thus posing a threat to the long-term stable development of coastal aquaculture areas.

[0003] In order to effectively deal with this problem, it is particularly important to accurately monitor and warn of the ocean environment temperature. However, some traditional marine environment monitoring methods rely on discrete sampling points and limited monitoring frequency, so it is difficult to fully and real-time reflect the dynamic changes of the marine environment. In addition, due to various errors and interference factors in the data collection and processing process, the monitoring accuracy and reliability of traditional methods also need to be improved. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for monitoring the environmental temperature of a marine area based on a coastal aquaculture area, which can provide timely and accurate environmental information for the coastal aquaculture area.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In the first aspect, a method for monitoring the environmental temperature of a marine area in a coastal aquaculture area comprises:

[0007] Obtain parameter data of the ocean environment and data at the edge nodes of the acquisition equipment through acquisition equipment;

[0008] Eliminate random errors in edge node data, identify and process outliers, missing values, and duplicate values ​​in the data to obtain processed data;

[0009] Fusing the processed data with parameter data of the ocean environment to obtain fused data;

[0010] Conduct multi-dimensional factor analysis on the fused data to obtain analysis results;

[0011] According to the analysis results, the parameter data of the marine environment are analyzed in terms of time-space correlation and attribute correlation to obtain marine time-space information;

[0012] Fuzzy division of ocean space-time information to obtain dynamic change characteristic data of the ocean environment;

[0013] By establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, the spatial gradient distribution characteristics of the ocean data can be obtained;

[0014] Based on the spatial gradient distribution characteristics, the changing trend of ocean ambient temperature can be identified to provide environmental warnings for coastal aquaculture areas.

[0015] Furthermore, random errors in edge node data are eliminated, and outliers, missing values, and duplicate values ​​in the data are identified and processed to obtain processed data, including:

[0016] Determine the maximum and minimum values ​​in the data set, set the detection threshold, and obtain the critical value for data verification;

[0017] pass The degree of deviation of each data point from the mean is obtained, where x i is the i-th data point in the dataset, n is the number of data points in the dataset, G i is the difference between each data point and the mean;

[0018] Compare the degree of deviation of the data point from the mean with the critical value of data verification to eliminate the random errors of edge node data;

[0019] According to the processed edge node data, outliers, missing values ​​and duplicate values ​​in the data are processed to obtain processed data.

[0020] Furthermore, the processed data is fused with the parameter data of the ocean environment to obtain fused data, including:

[0021] Depending on the different acquisition equipment, the processed data and the parameter data of the marine environment are regarded as different bodies of evidence to obtain each body of evidence;

[0022] Use historical data to assign an initial probability distribution value to each body of evidence, and combine multiple bodies of evidence according to the probability distribution value to obtain the final decision value;

[0023] The final decision value is weighted spatially and temporally to obtain the fused data.

[0024] Furthermore, the final decision value is weighted spatially and temporally to obtain fused data, including:

[0025] According to the final decision value, the weighting coefficient of the time dimension and the weighting coefficient of the space dimension are set, and the weighting coefficient is calculated by To determine the spatiotemporal weighted value, where t si is the measurement time of the ith sensor, tf is the calibration time, |t si -t f | is the difference between the measurement time and the calibration time of sensor i, a is the weighting coefficient of the time dimension, b is the weighting coefficient of the space dimension, lon si ,lat si is the longitude and latitude of the geographical location of the ith sensor, lon f ,lat f is the longitude and latitude of the location of the calibration point, R is the radius of the earth, s i is the spatiotemporal weighted value of the i-th sensor;

[0026] According to the time-space weighted value, Calculate the space-time offset distance, where is the average of the spatiotemporal weighted values ​​of all sensors, d j is the temporal and spatial offset distance of the jth sensor, N is the number of sensors, and s j is the spatiotemporal weighted value of the jth sensor;

[0027] According to the time-space offset distance, Weighted fusion data is used to obtain fused data s, where is the confidence of the jth sensor, i and j are indices.

[0028] Furthermore, multi-dimensional factor analysis is performed on the fused data to obtain analysis results, including:

[0029] Perform principal component analysis on the fused data to obtain the environmental data trend in the data;

[0030] The fused data set is layered in the vertical space, and in the horizontal space, the data set is divided into several regions. In time, the combined time step is set, and the spatiotemporal regions are clustered and classified to obtain classified data.

[0031] Furthermore, according to the analysis results, the parameter data of the marine environment are analyzed for spatiotemporal correlation and attribute correlation to obtain marine spatiotemporal information, including:

[0032] Based on the analysis results, a spatiotemporal database of marine environmental parameters is constructed;

[0033] According to the spatiotemporal database, the marine environmental parameters of the unobserved areas are estimated by inverse distance weighted interpolation to obtain the spatiotemporal distribution data of the marine environmental parameters;

[0034] According to the spatiotemporal distribution data of marine environmental parameters, the changing trends of marine environmental parameters in different time periods and different spatial locations are analyzed to obtain spatiotemporal analysis results;

[0035] Analyze the correlation between marine environmental parameters to obtain correlation analysis results;

[0036] According to the correlation analysis results, the marine environmental parameters with similar attributes are grouped to obtain the attribute association analysis results.

[0037] Furthermore, the ocean space-time information is fuzzily divided to obtain the dynamic change characteristic data of the ocean environment, including:

[0038] According to the ocean spatiotemporal information, the membership threshold and the number of clusters are set, and the spatiotemporal data are fuzzy processed to obtain the fuzzy division results of each parameter;

[0039] According to the fuzzy division results, the areas where key parameters have changed significantly are identified, and the characteristic data related to the dynamic changes are extracted from the changed areas to obtain the dynamic change characteristic data of the marine environment.

[0040] The second aspect is the marine area environmental temperature monitoring system based on coastal aquaculture areas, including:

[0041] An acquisition module is used to acquire parameter data of the ocean environment and data at edge nodes of the acquisition device through an acquisition device;

[0042] The fusion module is used to eliminate random errors in edge node data, identify and process abnormal values, missing values ​​and duplicate values ​​in the data to obtain processed data; and fuse the processed data with the parameter data of the marine environment to obtain fused data;

[0043] The analysis module is used to perform multi-dimensional factor analysis on the fused data to obtain analysis results; based on the analysis results, the parameter data of the marine environment is analyzed for spatiotemporal correlation and attribute correlation to obtain marine spatiotemporal information;

[0044] The processing module is used to fuzzily divide the ocean's spatiotemporal information to obtain the dynamic change characteristic data of the ocean environment; by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, the spatial gradient distribution characteristics of the ocean data can be obtained; based on the spatial gradient distribution characteristics, the changing trend of the ocean environment temperature can be identified to provide environmental warnings for coastal aquaculture areas.

[0045] According to a third aspect, a computing device includes:

[0046] one or more processors;

[0047] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0048] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0049] The above solution of the present invention includes at least the following beneficial effects:

[0050] By eliminating random errors in edge node data, identifying and processing outliers, missing values ​​and duplicate values ​​in the data, the accuracy and reliability of the data can be significantly improved; fusing the processed data with the parameter data of the marine environment can make comprehensive use of multiple data sources, improve the amount of information and representativeness of the data, and make the analysis results more comprehensive and in-depth. Multidimensional factor analysis of the fused data helps to reveal the complex relationships and potential trends in the marine environment; through the spatiotemporal correlation and attribute correlation analysis of the analysis results, we can have a deeper understanding of the dynamic changes and interaction mechanisms of the marine environment; fuzzy division of marine spatiotemporal information helps to capture the dynamic change characteristics of the marine environment, and by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, we can reveal the spatial gradient distribution characteristics of marine data, which helps to identify sensitive areas and key processes in the marine environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of a method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area provided by an embodiment of the present invention.

[0052] Figure 2 It is a schematic diagram of a marine area environmental temperature monitoring system based on a coastal aquaculture area provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in a form and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0054] like Figure 1 As shown, an embodiment of the present invention proposes a method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area, and the method comprises the following steps:

[0055] Step 11, acquiring parameter data of the ocean environment and data at the edge nodes of the acquisition device through the acquisition device;

[0056] Step 12, eliminating random errors in edge node data, identifying and processing outliers, missing values, and duplicate values ​​in the data to obtain processed data;

[0057] Step 13, fusing the processed data with the parameter data of the ocean environment to obtain fused data;

[0058] Step 14, performing multidimensional factor analysis on the fused data to obtain analysis results;

[0059] Step 15, based on the analysis results, performing spatiotemporal correlation and attribute correlation analysis on the parameter data of the marine environment to obtain marine spatiotemporal information;

[0060] Step 16, fuzzy division of ocean space-time information to obtain dynamic change characteristic data of the ocean environment;

[0061] Step 17, by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, the spatial gradient distribution characteristics of the ocean data are obtained; according to the spatial gradient distribution characteristics, the change trend of the ocean environment temperature is identified to provide environmental warning for coastal aquaculture areas.

[0062] In the embodiment of the present invention, step 11 can collect the key parameters of the marine environment in real time and accurately. Step 12, through data cleaning and preprocessing, the quality and accuracy of the data are effectively improved. Step 13, data fusion can integrate various information, provide more comprehensive and accurate marine environmental data, and help reveal the true state of the marine environment. Step 14, multidimensional factor analysis can deeply mine the information in the data from multiple angles, and help to more comprehensively understand the characteristics and changing laws of the marine environment. Step 15, through spatiotemporal association and attribute association analysis, the inherent connection and mutual influence between marine environmental parameters can be revealed. Step 16, fuzzy division can better deal with the uncertainty and ambiguity in the marine environment, so as to more accurately capture the dynamic change characteristics of the marine environment. Step 17, by establishing a regular grid and calculating the spatial gradient distribution characteristics, the spatial distribution and change trend of the marine environmental temperature can be more intuitively displayed, providing timely and effective environmental warnings for coastal aquaculture areas, helping breeders to make more reasonable decisions, and reducing losses caused by changes in environmental temperature. At the same time, this method also helps to improve the efficiency and accuracy of marine environmental monitoring.

[0063] Preferably, step 11, obtaining the parameter data of the ocean environment and the data at the edge node of the acquisition device through the acquisition device, includes:

[0064] The parameter data of the ocean environment and the data at the edge nodes of the collection equipment are obtained through the collection equipment, and the data include temperature, salinity and flow rate.

[0065] In the embodiments of the present invention, real-time acquisition of temperature, salinity and flow rate data is helpful for continuous monitoring of the marine environment, and can timely detect abnormal changes, such as sudden temperature rise or abnormal salinity, so as to provide early warning of possible marine disasters or pollution incidents; understanding the temperature, salinity and flow rate information of a specific area is helpful for more effective management of marine resources. By analyzing these data, polluted or ecologically damaged areas can be identified, providing target areas and scientific basis for environmental protection and restoration work. Long-term recording of marine environmental parameter data helps to reveal the impact of climate change on the ocean.

[0066] Preferably, step 12, eliminating random errors in edge node data, identifying and processing outliers, missing values, and duplicate values ​​in the data to obtain processed data, includes:

[0067] Determine the maximum and minimum values ​​in the data set, set the detection threshold, and obtain the critical value for data verification;

[0068] pass The degree of deviation of each data point from the mean is obtained, where x i is the i-th data point in the dataset, n is the number of data points in the dataset, G i is the difference between each data point and the mean, i is the index;

[0069] Compare the degree of deviation of the data point from the mean with the critical value of data verification to eliminate the random errors of edge node data;

[0070] According to the processed edge node data, outliers, missing values ​​and duplicate values ​​in the data are processed to obtain processed data.

[0071] In an embodiment of the present invention, by identifying and processing outliers, missing values, and duplicate values, the noise and errors in the data set will be significantly reduced, thereby improving the overall data quality. Setting detection thresholds and eliminating random errors in edge node data helps to ensure the consistency and accuracy of the data. After removing invalid and erroneous data, the data set becomes more streamlined, reducing the computational burden during analysis, improving the efficiency of data processing and analysis, and making decisions based on high-quality data more reasonable and credible, thereby reducing the risk of decision-making errors caused by data problems. The processed data is easier to visualize and can more clearly display data features and trends. Performing these operations in the early stages of data processing can avoid repeated corrections and adjustments to the data in subsequent processes.

[0072] In a specific embodiment of the present invention, the entire data set is traversed to find the maximum value (maximum value) and the minimum value (minimum value) therein, which represent the range of values ​​in the data set. A threshold is set based on the standard deviation of the data set to determine whether a data point is abnormal. The critical value is obtained by adding or subtracting several times the standard deviation from the mean of the data set to divide the boundary between normal data and abnormal data. The degree of deviation of each data point from the mean is calculated using a formula to calculate the mean of the data set. For each data point x in the dataset i , calculate the absolute difference between it and the mean, and adjust this difference using the weight factor in the formula to obtain the degree of deviation G of each data point i , where x i is the i-th data point in the data set, n is the number of data points in the data set, and the first part of the formula is a weighting factor based on the sum of the squared deviations of all data points from the mean; the second part is the absolute difference between the current data point and the mean. i Compare with the previously set critical value. If the deviation of a data point exceeds the critical value, it may be regarded as an outlier. Based on the comparison results, identify the edge node data with excessive deviation due to random errors, identify abnormal data points in the data set, and replace (such as using interpolation, mean, etc.), delete, or retain but mark the data points as abnormal. Identify missing values ​​(usually null values ​​or specific marks) in the data set, and fill (using mean, median, mode, etc.), interpolate, or delete records containing missing values. Check whether there are duplicate records or data points in the data set, delete duplicates, or keep the first one that appears.

[0073] Preferably, step 13, fusing the processed data with the parameter data of the ocean environment to obtain fused data, includes:

[0074] Depending on the different acquisition equipment, the processed data and the parameter data of the marine environment are regarded as different bodies of evidence to obtain each body of evidence;

[0075] Use historical data to assign an initial probability distribution value to each body of evidence, and combine multiple bodies of evidence according to the probability distribution value to obtain the final decision value;

[0076] The final decision value is weighted spatially and temporally to obtain the fused data.

[0077] In an embodiment of the present invention, by treating the data acquired by different acquisition devices as different evidence bodies, the characteristics and accuracy of each data source can be considered more carefully, thereby improving the accuracy of data fusion. By using historical data to assign an initial probability distribution value to each evidence body, the credibility of each evidence body can be evaluated more scientifically, and then a more reliable final decision value can be obtained when merging multiple evidence bodies. By fusing data from different sources, multi-source information can be integrated to obtain a more comprehensive and accurate description of the marine environment, which helps to better understand and analyze the marine environment. By weighting the final decision value in space and time, it can better adapt to the dynamic changes of the marine environment, making the fused data more real-time and regional. Accurate data fusion helps to more reasonably allocate and utilize marine resources. The data processing method combines multidisciplinary knowledge such as probability theory, information fusion, and geographic information system, which helps to promote the cross-application and development of related disciplines. The fused data can provide more accurate and comprehensive input information for the decision support system, thereby improving the quality and efficiency of decision-making.

[0078] In a specific embodiment of the present invention, the body of evidence refers to a data set from a collection device that has been preliminarily processed (such as abnormal values, missing values, and duplicate values). Since different collection devices may have different accuracy, measurement ranges, and response characteristics, the data is divided into different groups according to the type of collection device (such as temperature sensor, salinity sensor, flow rate sensor, etc.). For each group of data, necessary preprocessing is performed, including data cleaning and format conversion. Each group of data after preprocessing constitutes an independent body of evidence. Based on the statistical analysis of historical data, an initial probability distribution value is assigned to each body of evidence. The probability distribution value reflects the credibility or weight of each body of evidence. For example, the data accuracy, stability, and other indicators of a certain device over a period of time can be analyzed to assign an initial probability value to it. After the initial probability distribution value is obtained, these bodies of evidence are merged by weighted averaging. During the merging process, the data of each body of evidence will be weighted according to its corresponding probability distribution value. The result of the merging is a final decision value that integrates the information of all the bodies of evidence, representing the best estimate of the marine environmental parameters under current conditions. Due to the spatial variability of the marine environment, data at different locations may have different importance or influence. Therefore, different spatial weights are determined by spatial statistical analysis methods (such as geographically weighted regression) based on the spatial location of the data point (such as longitude and latitude, water depth, etc.). Marine environmental parameters also usually change over time, and time weights are assigned based on the timestamp of the data point. After applying spatial and temporal weighting, the final decision value is adjusted through a formula to obtain more accurate and comprehensive fused data.

[0079] Preferably, step 13, weighting the final decision value in space and time to obtain fused data, includes:

[0080] According to the final decision value, the weighting coefficient of the time dimension and the weighting coefficient of the space dimension are set, and the weighting coefficient is calculated by

[0081] To determine the spatiotemporal weighted value, where t si is the measurement time of the ith sensor, t f is the calibration time, |t si -t f | is the difference between the measurement time and the calibration time of sensor i, a is the weighting coefficient of the time dimension, b is the weighting coefficient of the space dimension, lon si ,lat si is the longitude and latitude of the geographical location of the ith sensor, lon f ,lat f is the longitude and latitude of the location of the calibration point, R is the radius of the earth, s i is the spatiotemporal weighted value of the i-th sensor, i is the index;

[0082] According to the time-space weighted value, Calculate the space-time offset distance, where is the average of the spatiotemporal weighted values ​​of all sensors, d j is the temporal and spatial offset distance of the jth sensor, N is the number of sensors, and s j is the spatiotemporal weighted value of the jth sensor, i and j are indices;

[0083] According to the time-space offset distance, Weighted fusion data is used to obtain fused data, where: is the confidence of the jth sensor, i and j are indices.

[0084] In an embodiment of the present invention, by introducing weighting coefficients of the time dimension and the space dimension, the position of each sensor data in time and space can be more accurately located, thereby improving the time and space resolution and accuracy of the data. The time and space weighted value can dynamically reflect the difference between the sensor measurement time and the calibration time and the difference between the sensor geographical location and the calibration point location, so that the data fusion process can be more adaptable to changes in actual conditions. By calculating the time and space offset distance, the degree of deviation of each sensor data from the overall data set can be quantified, so that appropriate weights can be given during data fusion to optimize the fusion effect. The weighted fusion data method can reduce the impact of abnormal data from individual sensors on the overall data and enhance the robustness and stability of the data. By performing time and space weighted processing on sensor data, limited resources can be allocated and utilized more reasonably.

[0085] In a specific embodiment of the present invention, a is the weighting coefficient of the time dimension, and b is the weighting coefficient of the space dimension, which is used to adjust the influence of the spatial factor on data fusion. For each sensor, first calculate the difference between its measurement time and calibration time |t si -t f |, reflects the temporal freshness of the data. Then, the formula of the spherical cosine law is used to calculate the spatial distance between the sensor and the calibration point; the time difference and spatial distance are multiplied by the corresponding weighting coefficients, and then added to obtain the spatiotemporal weighted value of each sensor. First, the average of the spatiotemporal weighted values ​​of all sensors is calculated. The average of the spatiotemporal weighted values ​​represents the overall central trend of the sensor data. For each sensor, the absolute difference between its spatiotemporal weighted value and the average value is calculated and divided by the standard deviation of all sensor weighted values ​​(i.e., the square root of the variance). The obtained d j Indicates the temporal and spatial offset of the jth sensor data relative to the overall data. Use the temporal and spatial offset distance d j As the basis of fusion weights. Sensor data with larger offset distances will be given greater weights during fusion. The final fusion data is obtained by multiplying the spatiotemporal weighted value of each sensor with its corresponding offset distance weight and summing them.

[0086] Preferably, step 14, performing multi-dimensional factor analysis on the fused data to obtain analysis results, includes:

[0087] Perform principal component analysis on the fused data to obtain the environmental data trend in the data;

[0088] The fused data set is layered in the vertical space, and in the horizontal space, the data set is divided into several regions. In time, the combined time step is set, and the spatiotemporal regions are clustered and classified to obtain classified data.

[0089] In an embodiment of the present invention, principal component analysis can extract the main change trends in the data and reduce the multidimensional data to a few principal components, thereby simplifying the structure of the data; through principal component analysis, the most important environmental variables and their change trends in the fused data can be identified. After the data is reduced in dimension, the amount of calculation is greatly reduced, which accelerates the speed of data processing and analysis, making real-time monitoring and rapid response possible. By layering in vertical space, dividing regions in horizontal space, and setting time steps, the spatiotemporal division method can more accurately capture the data characteristics in different regions and time periods. Clustering classification of spatiotemporal regions helps to discover hidden patterns and group structures in the data, such as regions or time periods with similar environmental conditions. Based on classified data, monitoring resources can be allocated more reasonably. After understanding the main trends and spatiotemporal classification characteristics of environmental data, the ability to predict future environmental changes can be improved.

[0090] In a specific embodiment of the present invention, principal component analysis is a commonly used data dimension reduction technology, which aims to extract the main change trends in the data. By performing principal component analysis on the fused data, the most important components in the data set can be identified. The data is standardized to eliminate the influence of different dimensions and orders of magnitude on the data analysis results. Standardization can be achieved by subtracting the mean and dividing by the standard deviation, so that the mean of each variable is 0 and the standard deviation is 1; the covariance matrix reflects the correlation between different variables. For the standardized data, its covariance matrix is ​​calculated. The covariance matrix is ​​subjected to eigenvalue decomposition to obtain eigenvalues ​​and corresponding eigenvectors, wherein the size of the eigenvalue represents the degree of change of the data in the direction of the corresponding eigenvector, that is, the size of the variance. According to the size of the eigenvalue, the eigenvectors corresponding to the first few larger eigenvalues ​​are selected as the principal components. The principal components can retain most of the information of the original data and realize the dimensionality reduction of the data. The original data is linearly transformed using the selected principal components (i.e., eigenvectors) to obtain a new data set, which expresses the main change trends of the original data in the principal component space. According to the vertical distribution characteristics of the data (such as water depth, temperature gradient, etc.), the data set is divided into several layers in the vertical direction, and each layer represents similar vertical environmental conditions. In the horizontal direction, according to factors such as geographical location and environmental characteristics, the data set is divided into several regions. These regions can be defined based on grid division, natural geographical boundaries or other relevant standards. In the time dimension, a suitable time step is set (such as day, week, month, etc.), and the data is aggregated according to this time step. Aggregation can be achieved by calculating the mean, median or other statistics within each time step. Combining the data after vertical stratification, horizontal region division and time aggregation, a multidimensional spatiotemporal data set is formed. The K-means clustering algorithm is applied to cluster analysis of this data set. The clustering results divide the spatiotemporal regions into different categories, and each category represents similar environmental data characteristics and change trends.

[0091] Preferably, step 15, based on the analysis results, performs spatiotemporal correlation and attribute correlation analysis on the parameter data of the ocean environment to obtain ocean spatiotemporal information, including:

[0092] Based on the analysis results, a spatiotemporal database of marine environmental parameters is constructed;

[0093] According to the spatiotemporal database, the marine environmental parameters of the unobserved areas are estimated by inverse distance weighted interpolation to obtain the spatiotemporal distribution data of the marine environmental parameters;

[0094] According to the spatiotemporal distribution data of marine environmental parameters, the changing trends of marine environmental parameters in different time periods and different spatial locations are analyzed to obtain spatiotemporal analysis results;

[0095] Analyze the correlation between marine environmental parameters to obtain correlation analysis results;

[0096] According to the correlation analysis results, the marine environmental parameters with similar attributes are grouped to obtain the attribute association analysis results.

[0097] In an embodiment of the present invention, a spatiotemporal database of marine environmental parameters is constructed, which can realize the systematic storage and management of data, facilitate the rapid query and retrieval of historical and real-time data, and improve the convenience and efficiency of data use. Estimating the marine environmental parameters of unobserved areas by inverse distance weighted interpolation can fill the data gaps, improve the understanding of the distribution of environmental parameters in the entire sea area, and help predict the environmental conditions of unobserved areas. Spatiotemporal distribution data can intuitively show the changing trends of marine environmental parameters in different time periods and different spatial locations, and statistically analyze the correlation between marine environmental parameters to reveal the mutual influence and dependency between different parameters. Using the K-means algorithm to group marine environmental parameters with similar attributes helps to identify parameter groups with similar characteristics, and then more targeted monitoring and management strategies can be formulated for different groups.

[0098] In a specific embodiment of the present invention, based on the analysis results, a spatiotemporal database of marine environmental parameters is constructed, which may specifically include:

[0099] Collect data on marine environmental parameters from various sources, such as temperature, salinity, dissolved oxygen, pH value, and nutrient concentration. These data can come from field observations, remote sensing monitoring, model predictions, etc. Clean the collected data, remove outliers, duplicate values, and records with missing key information, format the data, and ensure that all data follow a unified format and standard. Design the structure of the spatiotemporal database, including data tables, fields, and associations, and use the timestamp and spatial coordinates of the data as key fields. Import the preprocessed data into the spatiotemporal database, establish appropriate indexes, and build a spatiotemporal database of marine environmental parameters.

[0100] In a specific embodiment of the present invention, based on the spatiotemporal database, the ocean environment parameters of the unobserved area are estimated by inverse distance weighted interpolation to obtain the spatiotemporal distribution data of the ocean environment parameters, which may specifically include:

[0101] pass Determine the search radius or the number of neighbor points, where z(m,y) is the ocean environment parameter (such as temperature, salinity, etc.) of the point to be estimated (m,y), n1 is the total number of data points, and z i is the ocean environment parameter value at the known data point i, (m,y) is the coordinate of the point to be estimated, (m i ,y i) is the coordinate of the known data point i, p is the weight index, β is the weight adjustment parameter, γ is the time decay parameter, is the time decay term, which means that the greater the time difference, the smaller the impact; η(m,y) is the spatial heterogeneity parameter, and i is the index. To define the range of observation points considered during interpolation, set the distance decay function, where the distance decay function is the reciprocal of the distance or the square of the reciprocal. For each unobserved point, calculate the spatial distance between it and all known observation points, which can be done using the Euclidean distance formula. The above distance decay function is used to calculate the weight of each observation point to the unobserved point. The closer the distance, the greater the weight; the farther the distance, the smaller the weight. The ocean environment parameter value of each observation point is multiplied by its corresponding weight to obtain the contribution value of the observation point to the unobserved point (wherein, before weight multiplication, the ocean environment parameter value of each observation point is first standardized. Standardization is to subtract the parameter value from its mean value and then divide it by its standard deviation, so as to convert the parameter value to the same scale. For example, the mean is 0 and the standard deviation is 1. The standardized parameter value is multiplied by its corresponding weight to calculate the contribution value of each observation point to the unobserved point. Since the parameter value is now on the same scale, the multiplication operation will not be biased due to the different original scales of the parameter). The contribution values ​​of all observation points to the unobserved point are added together to obtain the total weighted contribution. At the same time, the sum of the weights of all observation points is calculated. The total weighted contribution is divided by the sum of weights to obtain the estimated value of the ocean environment parameter of the unobserved point. This process realizes weighted averaging and ensures that observation points at different distances make reasonable contributions to the value of the unobserved point according to their weights. The parameter estimates of all unobserved points obtained through the above interpolation process are combined with their corresponding geographic spatial locations. For example, geographic information system (GIS) software can be used to map the parameter estimates of all unobserved points to geographic space to generate spatiotemporal distribution data of marine environmental parameters, so that the distribution of marine environmental parameters in different time and space can be intuitively displayed.

[0102] In a specific embodiment of the present invention, according to the spatiotemporal distribution data of the marine environmental parameters, the changing trends of the marine environmental parameters in different time periods and different spatial positions are analyzed to obtain spatiotemporal analysis results, which specifically include:

[0103] Collect long-term and continuous marine environmental parameter data from the spatiotemporal database to ensure that the data set contains sufficient time span and frequency to capture the long-term change trend of the parameters; preprocess the data, including data cleaning, outlier detection and processing, missing value interpolation, etc., to ensure the quality and reliability of the data; select independent variables (such as time, season, climate factors, etc.) and dependent variables (i.e., the marine environmental parameters to be predicted) according to the research purpose and the characteristics of the marine environmental parameters; construct a linear regression analysis model using the selected independent and dependent variables; fit the constructed linear regression analysis model using historical data, and estimate the parameters of the linear regression analysis model by the least squares method; evaluate the fitted linear regression analysis model, including checking the residuals of the model, calculating the explanatory indicators of the linear regression analysis model (such as the R-square value), and performing cross-validation, etc., to ensure the accuracy and generalization ability of the model; analyze the long-term change trend of the marine environmental parameters based on the fitted regression model, and by observing the coefficients and trend terms in the model, you can understand the direction and rate of change of the parameters over time. The linear regression analysis model is used to predict parameter changes in the future. This can be achieved by substituting the future independent variable values ​​into the linear regression analysis model and calculating the corresponding dependent variable prediction values.

[0104] In another specific embodiment of the present invention, analyzing the correlation between the marine environment parameters to obtain the correlation analysis result may include:

[0105] The relevant parameter data to be analyzed are extracted from the spatiotemporal database, and the linear relationship strength and direction between the marine environmental parameters are measured by the Pearson correlation coefficient. Calculate the correlation coefficient between each pair of parameters and generate a correlation coefficient matrix, where r ij is the Pearson correlation coefficient between the i-th and j-th parameters (variables), X it and X jt are the observed values ​​of the i-th and j-th parameters in the t-th sample, and are the means of the i-th and j-th parameters respectively, t is the index, and n2 is the total number of data points.

[0106] In a specific embodiment of the present invention, according to the correlation analysis result, the marine environment parameters with similar attributes are grouped to obtain the attribute association analysis result, which may include:

[0107] Perform correlation analysis between marine environmental parameters, such as using the Pearson correlation coefficient method; according to the results of the correlation analysis, set thresholds to automatically identify parameter pairs with strong, weak or no correlation. For example, a correlation coefficient greater than 0.7 can be considered a strong correlation, between 0.3 and 0.7 is a weak correlation, and less than 0.3 is no correlation. From the results of the correlation analysis, select representative features that are not strongly correlated with each other for subsequent clustering analysis to avoid information redundancy; perform necessary preprocessing on the selected features, such as standardization and normalization, to ensure that they have the same weight in the clustering process.

[0108] Before performing K-means clustering, the optimal number of clusters K is determined by methods such as the Elbow Method or Silhouette Score. When initializing the cluster centers, the K-means algorithm can be used to optimize the selection of the initial centers to reduce the impact of the initial values ​​on the clustering results. During the iteration process, the Euclidean distance metric is used to assign data points to the nearest cluster centers, and the positions of the cluster centers are recalculated. The iteration process is repeated until the change in the cluster centers reaches the maximum number of iterations to obtain the clustering results. Based on the clustering results, the parameter composition in each cluster is analyzed to identify the dominant parameters and features. The identification can be assisted by calculating the mean, median or mode of each parameter in each cluster.

[0109] Preferably, step 16, fuzzy division of ocean spatiotemporal information to obtain dynamic change characteristic data of the ocean environment, includes:

[0110] According to the ocean spatiotemporal information, the membership threshold and the number of clusters are set, and the spatiotemporal data are fuzzy processed to obtain the fuzzy division results of each parameter;

[0111] According to the fuzzy division results, the areas where key parameters have changed significantly are identified, and the characteristic data related to the dynamic changes are extracted from the changed areas to obtain the dynamic change characteristic data of the marine environment.

[0112] In an embodiment of the present invention, by setting the membership threshold and the number of clusters, the spatiotemporal data can be flexibly fuzzy processed, the uncertainty and ambiguity in the data can be handled, and the complex and changeable situation in the marine environment can be better adapted. Based on the fuzzy division results, the areas where the key parameters have changed significantly can be identified more accurately. Extracting characteristic data related to dynamic changes from the changing areas helps to capture the dynamic change characteristics of the marine environment; by obtaining the dynamic change characteristic data of the marine environment, more accurate and timely information support can be provided to decision makers, improving the scientificity and effectiveness of decision-making. Dynamic change characteristic data provides empirical material for marine scientific research. By accurately identifying the areas where key parameters change, resource allocation can be optimized, such as adjusting the layout of monitoring sites, increasing the monitoring frequency of specific areas, etc.

[0113] In a specific embodiment of the present invention, a marine environmental parameter data set containing temporal and spatial information is collected and organized to ensure that the data set contains enough observation points to reflect the spatial variability and temporal dynamics of the marine environment. An appropriate fuzzy set is selected to describe each marine environmental parameter. The fuzzy set allows data points to belong to multiple categories with different degrees of membership. A membership function is defined for each parameter, which maps the parameter value to a membership between 0 and 1, wherein the membership represents the degree to which the data point belongs to a particular category, and a membership threshold is determined to divide whether the data point significantly belongs to a fuzzy set. For example, the threshold can be set to 0.5, indicating that the data point is considered to belong to the category only when the membership exceeds 0.5. The number of categories into which the data is to be clustered is determined based on prior knowledge, data visualization, or clustering effectiveness indicators. The fuzzy C-means algorithm is selected, and the data set, membership function, membership threshold and number of clusters are used as inputs to run the fuzzy clustering algorithm. The algorithm will iteratively optimize the membership of each data point to all categories until the stopping condition is met (such as reaching the maximum number of iterations or the membership change is less than the preset threshold). After the algorithm ends, the membership value of each data point to all categories is obtained. According to the membership threshold, the data points are divided into different fuzzy sets to obtain the fuzzy partitioning results of each parameter. From the numerous marine environmental parameters, the parameters that are critical to the research objectives are selected as key parameters. For each key parameter, its fuzzy partitioning results are checked, and special attention is paid to the data points whose membership values ​​are close to or exceed the preset threshold. These data points indicate that the parameter values ​​have changed significantly in time and space, representing important environmental processes or events. The data points with significant membership changes are located in geographic space, and these points are connected into regions through spatial analysis tools (such as GIS software) to identify continuous areas where key parameters have changed significantly. According to the research objectives, features related to dynamic changes are selected for extraction. These features may include the area, shape, location, time range, change rate, and change amplitude of the change area. For each feature, spatial analysis tools and statistical methods are used to extract selected feature data from the significant change area, and the extracted feature data are cleaned, sorted and quality checked. The correlation and interaction between the feature data are analyzed to deeply understand the dynamic change process of the marine environment.

[0114] Preferably, step 17, by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data to obtain the spatial gradient distribution characteristics of the ocean data, includes:

[0115] Divide the ocean area into grids to establish a regular grid;

[0116] Mapping dynamically changing feature data to a grid to calculate the average value within the grid cells;

[0117] According to the average value of the grid cells, the gradient is calculated to obtain the gradient distribution data;

[0118] Based on the gradient distribution data, the spatial variation patterns and trends of ocean data are identified to obtain the spatial gradient distribution characteristics of ocean data.

[0119] In an embodiment of the present invention, by dividing the ocean area into regular grids and calculating the average value of the characteristic data in each grid unit, a refined spatial analysis of the ocean environment can be achieved, and subtle changes in the ocean data in space can be captured more accurately. Gradient distribution data can clearly show the changing laws and trends of ocean data in space, which helps to quickly understand complex ocean environmental phenomena and improve decision-making efficiency. Through gradient calculation, the spatial gradient distribution characteristics of ocean data can be obtained. Gradient distribution data can help identify abnormal areas in the marine environment, such as abnormal diffusion of pollutants and abnormal changes in ecosystems. Understanding the spatial gradient distribution characteristics of ocean data can provide a scientific basis for the optimal allocation of marine resources. The refined spatial analysis data and intuitive visualization results provided can provide strong support for marine scientific research.

[0120] In a specific embodiment of the present invention, the size of the grid is determined according to the spatial resolution of the data. The grid size should be able to capture the spatial variation details of the marine environmental parameters while avoiding data redundancy caused by over-refinement. Use GIS software or programming tools (such as Python's GDAL / OGR library) to generate a regular grid system in the study area. Each grid cell should have a unique identifier (such as row number and column number), and the generated grid system is saved as a spatial data file. Use spatial connection or rasterization operation to map the dynamic change feature data to the grid cell. For each grid cell, collect all the dynamic change feature data points that fall within the cell. For the data points collected in each grid cell, calculate the arithmetic mean of its dynamic change characteristics, and use the calculated mean value as the representative value of the grid cell. Save the mean value of each grid cell as a new spatial data file or attribute table. According to the characteristics of the data and the purpose of analysis, the gradient of the average value of the dynamic change characteristics between it and the adjacent grid cells is calculated by the finite difference method or the central difference method. The size of the gradient indicates the severity of the change, and the direction of the gradient indicates the main trend of the change. Assign the calculated gradient value to the corresponding grid cell, and use GIS software or data visualization tools to generate gradient distribution data. Through the gradient distribution data, you can intuitively see the spatial variation patterns and trends of marine environmental parameters. Carefully check the gradient distribution data, pay attention to the size, direction and spatial distribution pattern of the gradient, and look for obvious gradient change areas, transition zones or outliers. According to the size and direction of the gradient, identify the main variation patterns and trends of marine environmental parameters. For example, you can identify areas where parameter values ​​increase or decrease, as well as the main direction of change (such as along the coast, perpendicular to the coastline, etc.). Combine professional knowledge and other auxiliary data (such as ocean currents, wind direction, etc.) to explain the causes and effects of these variation patterns and trends. Summarize and extract the spatial gradient distribution characteristics of marine environmental parameters, such as the size range of the gradient, the main direction of change, and spatial continuity, to obtain processed marine characteristic data.

[0121] In an embodiment of the present invention, according to the spatial gradient distribution characteristics, the change trend of the ocean environment temperature is identified to provide environmental warning for coastal aquaculture areas, specifically including:

[0122] Integrate historical ocean ambient temperature data with currently acquired gradient distribution characteristic data, including:

[0123] The specific steps for integrating historical ocean ambient temperature data with currently acquired gradient distribution characteristic data are as follows:

[0124] Obtain the historical marine ambient temperature dataset and the currently acquired gradient distribution characteristic dataset. These datasets are preprocessed and cleaned to ensure the quality and accuracy of the data; check the data formats of the two datasets to ensure that they have the same spatial and temporal resolution, or at least can be matched by interpolation and other methods; if the data format or resolution does not match, the data needs to be resampled, interpolated or aggregated to align them; if the historical temperature data and gradient distribution characteristic data are based on different spatial grids or coordinate systems, convert them to the same coordinate system; use GIS software or related tools for spatial transformation and registration to ensure that the spatial position of each data point is consistent; ensure that the timestamp of the historical temperature data can correspond to the timestamp of the current gradient distribution characteristic data. If the timestamps do not match, the data needs to be temporally interpolated to estimate the temperature value at the time point of the gradient data. Create a new dataset to store the integrated data, and for each spatial location and time point, merge the historical temperature data and gradient distribution characteristic data into the new dataset.

[0125] Preprocess the data, such as filling missing values ​​and processing outliers; arrange the integrated data in chronological order to construct a time series data set of ocean ambient temperature; smooth or seasonally adjust the data as needed to eliminate non-trend fluctuations; use time series analysis methods (such as moving average method) to identify and extract long-term trends in ocean ambient temperature; analyze the causes of long-term trends, such as climate change and ocean circulation; analyze the short-term fluctuation characteristics of ocean ambient temperature on the basis of removing long-term trends; use statistical methods (such as variance analysis) to evaluate the amplitude and frequency of short-term fluctuations and identify possible periodic changes.

[0126] Analyze the characteristics of historical data, including the stationarity, seasonality, and periodicity of the data, to determine whether the data needs to be differentiated or otherwise transformed to meet the requirements of the ARIMA model; based on the data characteristics, select the ARIMA model order (p, d, q), where p represents the order of the autoregressive term, d represents the order of the difference, and q represents the order of the moving average term. Divide the integrated historical marine ambient temperature data into a training set and a validation set. The training set is used for model training, while the validation set is used to evaluate the performance of the model; if the data is seasonal or periodic, ensure that the training set contains at least one complete cycle so that the model can learn these patterns. Use the training set data to train the selected ARIMA model, which involves estimating the parameters of the model, such as the autoregressive coefficient, the moving average coefficient, etc.; during the training process, use methods such as least squares and maximum likelihood estimation to optimize the model parameters; use the validation set data to evaluate the performance of the trained ARIMA model, and the evaluation indicators include mean square error (MSE), root mean square error (RMSE), etc. If the model performance is poor, adjust the order (p, d, q) of the ARIMA model or try other model configurations, then retrain and evaluate the model, and repeat this process until a model that can better fit the historical data and capture key trends is found. Once a satisfactory model is obtained, use the current data and historical trends as input to predict the trend of changes in the ocean ambient temperature for a period of time in the future. Generate forecast results, including forecast values ​​at various future time points and corresponding confidence intervals. The confidence interval can provide the uncertainty range of the forecast, which helps to understand the reliability of the forecast results. Compare the forecast results with the actual observations (if any) to verify the accuracy of the forecast, analyze the forecast trend, identify possible upward or downward trends, and any seasonal or periodic patterns, and formulate corresponding strategies or plans based on the forecast results to cope with future changes in ocean ambient temperature.

[0127] According to the sensitivity analysis of coastal aquaculture areas to marine environmental temperature, combined with extreme values ​​and abnormal events in historical data, reasonable warning thresholds are set. The warning thresholds include single thresholds or multiple graded thresholds to deal with different degrees of temperature change risks; deploy marine environmental monitoring systems to continuously monitor marine environmental temperature data in real time; compare real-time monitoring data with forecast results, and when the forecast results exceed the set warning threshold, immediately trigger the warning system to generate warning information, including the expected temperature change trend, possible impact range and recommended response measures, and send the warning information to managers and relevant stakeholders in coastal aquaculture areas through effective information transmission channels (such as SMS, email, special APP, etc.). After receiving the warning information, managers of coastal aquaculture areas make timely adjustments according to the recommended response measures to reduce or avoid the adverse effects of temperature changes on aquaculture activities, track the implementation effect of response measures, and make timely adjustments according to actual conditions.

[0128] like Figure 2 As shown, the embodiment of the present invention also provides a marine area environmental temperature monitoring system 20 based on a coastal aquaculture area, comprising:

[0129] An acquisition module 21 is used to acquire parameter data of the ocean environment and data at edge nodes of the acquisition device through an acquisition device;

[0130] The fusion module 22 is used to eliminate random errors in edge node data, identify and process abnormal values, missing values ​​and duplicate values ​​in the data to obtain processed data; and fuse the processed data with parameter data of the ocean environment to obtain fused data;

[0131] The analysis module 23 is used to perform multi-dimensional factor analysis on the fused data to obtain analysis results; based on the analysis results, perform spatiotemporal correlation and attribute correlation analysis on the parameter data of the marine environment to obtain marine spatiotemporal information;

[0132] The processing module 24 is used to perform fuzzy division on the ocean's spatiotemporal information to obtain dynamic change characteristic data of the ocean environment; to obtain the spatial gradient distribution characteristics of the ocean data by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data; and to identify the changing trend of the ocean environment temperature based on the spatial gradient distribution characteristics to provide environmental warnings for coastal aquaculture areas.

[0133] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area, characterized in that: The method comprises: Obtain parameter data of the ocean environment and data at the edge nodes of the acquisition equipment through acquisition equipment; Eliminate random errors in edge node data, identify and process outliers, missing values, and duplicate values ​​in the data to obtain processed data; Fusing the processed data with parameter data of the ocean environment to obtain fused data; Conduct multi-dimensional factor analysis on the fused data to obtain analysis results; According to the analysis results, the parameter data of the marine environment are analyzed in terms of time-space correlation and attribute correlation to obtain marine time-space information; Fuzzy division of ocean space-time information to obtain dynamic change characteristic data of the ocean environment; By establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, the spatial gradient distribution characteristics of the ocean data can be obtained; Based on the spatial gradient distribution characteristics, the changing trend of ocean ambient temperature can be identified to provide environmental warnings for coastal aquaculture areas.

2. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 1, characterized in that: Eliminate random errors in edge node data, identify and process outliers, missing values, and duplicate values ​​in the data to obtain processed data, including: Determine the maximum and minimum values ​​in the data set, set the detection threshold, and obtain the critical value for data verification; pass The degree of deviation of each data point from the mean is obtained, where x i is the i-th data point in the dataset, n is the number of data points in the dataset, G i is the difference between each data point and the mean; Compare the degree of deviation of the data point from the mean with the critical value of data verification to eliminate the random errors of edge node data; According to the processed edge node data, outliers, missing values ​​and duplicate values ​​in the data are processed to obtain processed data.

3. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 2, characterized in that: The processed data is fused with the parameter data of the ocean environment to obtain fused data, including: Depending on the different acquisition equipment, the processed data and the parameter data of the marine environment are regarded as different bodies of evidence to obtain each body of evidence; Use historical data to assign an initial probability distribution value to each body of evidence, and combine multiple bodies of evidence according to the probability distribution value to obtain the final decision value; The final decision value is weighted spatially and temporally to obtain the fused data.

4. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 3, characterized in that: The final decision value is weighted spatially and temporally to obtain fused data, including: According to the final decision value, the weighting coefficient of the time dimension and the weighting coefficient of the space dimension are set, and the weighting coefficient is calculated by To determine the spatiotemporal weighted value, where t si is the measurement time of the ith sensor, t f is the calibration time, |t si -t f | is the difference between the measurement time and the calibration time of sensor i, a is the weighting coefficient of the time dimension, b is the weighting coefficient of the space dimension, lon si ,lat si is the longitude and latitude of the geographical location of the ith sensor, lon f ,lat f is the longitude and latitude of the location of the calibration point, R is the radius of the earth, s i is the spatiotemporal weighted value of the i-th sensor; arccos is the inverse cosine function; According to the time-space weighted value, Calculate the space-time offset distance, where is the average of the spatiotemporal weighted values ​​of all sensors, d j is the temporal and spatial offset distance of the jth sensor, N is the number of sensors, and s j is the spatiotemporal weighted value of the jth sensor; According to the time-space offset distance, Weighted fusion data is used to obtain fused data s, where is the confidence of the jth sensor, i and j are indices.

5. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 4, characterized in that: Perform multidimensional factor analysis on the fused data to obtain analysis results, including: Perform principal component analysis on the fused data to obtain the environmental data trend in the data; The fused data set is layered in the vertical space, and in the horizontal space, the data set is divided into several regions. In time, the combined time step is set, and the spatiotemporal regions are clustered and classified to obtain classified data.

6. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 5, characterized in that: According to the analysis results, the parameter data of the marine environment are analyzed for spatiotemporal correlation and attribute correlation to obtain marine spatiotemporal information, including: Based on the analysis results, a spatiotemporal database of marine environmental parameters is constructed; According to the spatiotemporal database, the marine environmental parameters of the unobserved areas are estimated by inverse distance weighted interpolation to obtain the spatiotemporal distribution data of the marine environmental parameters; According to the spatiotemporal distribution data of marine environmental parameters, the changing trends of marine environmental parameters in different time periods and different spatial locations are analyzed to obtain spatiotemporal analysis results; Analyze the correlation between marine environmental parameters to obtain correlation analysis results; According to the correlation analysis results, the marine environmental parameters with similar attributes are grouped to obtain the attribute association analysis results.

7. The method for monitoring the environmental temperature of a marine area based on a coastal aquaculture area according to claim 6, characterized in that: Fuzzy division of ocean space-time information is performed to obtain dynamic change characteristic data of the ocean environment, including: According to the ocean spatiotemporal information, the membership threshold and the number of clusters are set, and the spatiotemporal data are fuzzy processed to obtain the fuzzy division results of each parameter; According to the fuzzy division results, the areas where key parameters have changed significantly are identified, and the characteristic data related to the dynamic changes are extracted from the changed areas to obtain the dynamic change characteristic data of the marine environment.

8. The marine area environmental temperature monitoring system based on coastal aquaculture areas is characterized by: Applied to the method according to any one of claims 1 to 7, comprising: An acquisition module is used to acquire parameter data of the ocean environment and data at edge nodes of the acquisition device through an acquisition device; The fusion module is used to eliminate random errors in edge node data, identify and process abnormal values, missing values ​​and duplicate values ​​in the data to obtain processed data; and fuse the processed data with the parameter data of the marine environment to obtain fused data; The analysis module is used to perform multi-dimensional factor analysis on the fused data to obtain analysis results; based on the analysis results, the parameter data of the marine environment is analyzed for spatiotemporal correlation and attribute correlation to obtain marine spatiotemporal information; The processing module is used to fuzzily divide the ocean's spatiotemporal information to obtain the dynamic change characteristic data of the ocean environment; by establishing the spatial distribution of the regular grid average scattered point dynamic change characteristic data, the spatial gradient distribution characteristics of the ocean data can be obtained; based on the spatial gradient distribution characteristics, the changing trend of the ocean environment temperature can be identified to provide environmental warnings for coastal aquaculture areas.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images