Marine environment real-time monitoring and early warning method and system

Abstract

The invention discloses a real-time monitoring and early warning system and method for a marine environment, and relates to the field of marine environment monitoring and early warning. Performing classification operation on the marine environment data in the historical marine environment data feature matrix by using a clustering analysis algorithm to obtain a marine environment category center data matrix, and setting an abnormal marine environment data category set; the method comprises the following steps: deploying a sensor in a marine environment, collecting real-time marine environment data, and preprocessing the real-time marine environment data to obtain preprocessed real-time marine environment data; extracting features from the preprocessed real-time marine environment data to construct a real-time marine environment data feature set, and calculating to obtain a real-time marine environment data type; and judging whether to send out early warning according to the real-time marine environment data type.

Description

Technical Field

[0001] The present invention belongs to the field of marine environment monitoring and early warning, and in particular, relates to a method and system for real-time monitoring and early warning of the marine environment. Background Art

[0002] With the intensification of global climate change and human activities, marine environmental problems are becoming increasingly serious, such as marine pollution, red tides, and rising sea levels. In order to effectively respond to these challenges, a real-time monitoring and early warning system for the marine environment has emerged. The system monitors marine environmental parameters in real time, such as water quality, meteorology, hydrology, etc., and combines artificial intelligence, big data and other technologies to analyze and predict changes in the marine environment, and issue early warning information in a timely manner, providing a scientific basis for marine environmental protection and disaster response.

[0003] Although the real-time monitoring and early warning system of the marine environment has made significant progress, there are still some shortcomings:

[0004] Insufficient data processing capabilities: The amount of marine environmental data is large and the types are complex. Existing data processing and analysis technologies cannot efficiently process and analyze massive amounts of data.

[0005] Insufficient data analysis capabilities; the accuracy and reliability of data analysis models need to be improved, making it difficult to accurately predict changing trends in the marine environment.

[0006] Imperfect early warning response mechanism: The response mechanism for early warning information is imperfect, making it difficult to effectively guide relevant departments and personnel to take response measures.

[0007] In view of the problems in the related technology, the present invention proposes a real-time monitoring and early warning method and system for the marine environment to overcome the above-mentioned technical problems existing in the existing related technology.

[0008] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0009] S1. Collect historical marine environment data, extract features, construct a feature matrix of historical marine environment data, use a cluster analysis algorithm to classify the marine environment data in the feature matrix of historical marine environment data, obtain a central data matrix of marine environment types, and set a set of abnormal marine environment data types according to the central data matrix of marine environment types;

[0010] S2. Deploy sensors in the marine environment, transmit real-time marine environment data to a central monitoring system through wireless transmission technology, pre-process the real-time marine environment data, and obtain pre-processed real-time marine environment data;

[0011] S3, extracting features from the pre-processed real-time marine environment data to construct a real-time marine environment data feature set, calculating the Euclidean distance from each feature data in the real-time marine environment data feature set to the corresponding type center data in the marine environment type center data matrix, obtaining a real-time marine environment data feature Euclidean distance set, taking the type corresponding to the smallest Euclidean distance in the real-time marine environment data feature Euclidean distance set as the type of the real-time marine environment data, and obtaining the real-time marine environment data type;

[0012] S4. Determine whether to issue an early warning based on the real-time ocean environment type.

[0013] Preferably, the S1 comprises the following steps:

[0014] S11, collect a large amount of historical marine environmental data, extract feature data, and construct the historical marine environmental data feature matrix A, as follows:

[0015]

[0016] Among them, A ib represents the bth feature of the i-th marine environmental data in the historical marine environmental data feature matrix, and a represents that there are a pieces of marine environmental data in the historical marine environmental data feature matrix;

[0017] S12, using a cluster analysis algorithm to classify the marine environment data of the historical marine environment data feature matrix to obtain a marine environment type center data matrix B, as follows:

[0018]

[0019] Among them, B ib represents the central data of all features b in the ith category in the central data matrix of marine environment categories, and c represents the total number of marine environment data categories in the central data matrix of marine environment categories;

[0020] S13, according to the marine environment type center data matrix B, set the abnormal marine environment data type set k = {k1, k2, ..., k i ,...,k l}, where k i represents the ith abnormal marine environment type in the abnormal marine environment data type set, l represents the total number of abnormal marine environment types in the abnormal marine environment data type set;

[0021] By collecting a large amount of historical marine environmental data and extracting characteristic data, constructing a historical marine environmental data characteristic matrix, using a cluster analysis algorithm to classify the marine environmental data, obtaining a central data matrix of marine environmental types, and setting an abnormal marine environmental data type set, it is helpful to identify and warn of abnormal marine environmental data, thereby better understanding the characteristics and changing laws of the marine environment, and providing an important basis for subsequent warnings.

[0022] Preferably, the S12 comprises the following steps:

[0023] S121, randomly selecting several pieces of historical marine environment data from the historical marine environment data feature matrix as initial category centers;

[0024] S122, using the Euclidean distance formula to calculate the distance from each piece of historical marine environmental data in the historical marine environmental data feature matrix to each category center of the initial category center; the Euclidean distance formula is as follows:

[0025] Among them, p is any data in the historical marine environmental data feature matrix, q is initially the species center, and p i and q i represents the i-th feature of the p-th and q-th historical marine environmental data;

[0026] Assign each piece of historical marine environmental data in the historical marine environmental data feature matrix to the species where the species center closest to it is located;

[0027] Calculate the average value of all historical marine environmental data in each category as the new category center; the calculation formula of the average value is as follows:

[0028]

[0029] Among them, C j represents the new center of category j, D j represents the number of historical marine environmental data in category j, E ij represents the value of the jth feature of the i-th historical marine environmental data in category j;

[0030] S123, iterating step S122, setting the maximum number of iterations to e1, the current number of iterations to e2, and the minimum convergence threshold to f;

[0031] When the change of the species center is ≤f or the previous iteration number is e2 ≥ the maximum iteration number is e1, the iteration operation is stopped to obtain the marine environment species center data matrix;

[0032] By randomly selecting initial species centers from the historical marine environmental data feature matrix and using the Euclidean distance formula to calculate the distance from each data to each species center, the data is assigned to the species where the nearest species center is located; by calculating the average value of all data in each species as the new species center, and iterating the process until the change in the species center is less than the minimum convergence threshold or the maximum number of iterations is reached, the marine environment species center data matrix is ​​obtained; it can effectively classify similar data together, revealing the potential patterns or structures in the marine environmental data, so as to better understand the characteristics and changing laws of the marine environment.

[0033] Preferably, S2 comprises the following steps:

[0034] S21. Select appropriate sensor types; determine the number, location and distribution of sensors according to the characteristics and requirements of the marine environment; install the sensors in the marine environment;

[0035] S22. According to the characteristics and requirements of the marine environment, select appropriate wireless transmission technology; establish a wireless transmission network, and transmit real-time data to the central monitoring system through the wireless transmission network to obtain real-time marine environment data;

[0036] S23, preprocessing the real-time ocean environment data to obtain preprocessed real-time ocean environment data;

[0037] By selecting appropriate sensor types and determining the number, location and distribution of sensors according to the characteristics and needs of the marine environment, and installing sensors in the marine environment, real-time marine environmental data can be effectively collected; by selecting appropriate wireless transmission technology and establishing a wireless transmission network, data can be transmitted to the central monitoring system in real time to obtain pre-processed real-time marine environmental data; changes in the marine environment can be monitored in real time, providing an important basis for environmental protection, prediction and warning, scientific research and international cooperation, so as to better understand the characteristics and changing laws of the marine environment.

[0038] Preferably, the S23 comprises the following steps:

[0039] S231, for the missing data in the real-time marine environment data, fill them with interpolation and mean filling methods, identify and process outliers, delete outliers, and delete duplicate data records; obtain the cleaned real-time marine environment data;

[0040] S232, using a low-pass filter to remove high-frequency noise from the cleaned real-time ocean environment data, retaining low-frequency signals, decomposing the data using wavelet transform to remove noise, retaining useful information, and using linear regression or polynomial regression to remove trend items in the data to obtain denoised real-time ocean environment data;

[0041] S233, performing interpolation and normalization processing on the denoised real-time ocean environment data to obtain pre-processed real-time ocean environment data;

[0042] These preprocessing steps help improve the reliability and usability of data, providing a solid foundation for subsequent data analysis and decision-making.

[0043] Preferably, S3 comprises the following steps:

[0044] S31, extracting features from the pre-processed real-time ocean environment data, and constructing a real-time ocean environment data feature set g = {g1, g2, ..., g i ,...,g b}, where g i represents the i-th feature of the real-time ocean environment data in the real-time ocean environment data feature set;

[0045] S32, the real-time ocean environment data feature set g={g1,g2,...,g i ,...,g b} to the corresponding type center data in the marine environment type center data matrix B, and obtain the real-time marine environment data feature Euclidean distance set h = {h1,h2,...,h i ,...,h c}, where h i Represents the Euclidean distance from the real-time marine environment data to the i-th species center data in the marine environment species center data matrix B;

[0046] S33, taking the real-time ocean environment data characteristic Euclidean distance set h={h1,h2,...,h i ,...,h c}, the type corresponding to the smallest Euclidean distance is taken as the type of real-time ocean environment data, and the type of real-time ocean environment data is obtained;

[0047] Through the above steps, the real-time marine environment data can be classified into the marine environment type that is most similar to it, so as to better understand the characteristics and changing laws of the marine environment, and provide important basis for subsequent judgment on whether to issue an early warning.

[0048] Preferably, S4 comprises the following steps:

[0049] S41, searching the type of real-time ocean environment data in the abnormal ocean environment data type set, and if the type of real-time ocean environment data belongs to the type in the abnormal ocean environment data type set, issuing an early warning; if the type of real-time ocean environment data does not belong to the type in the abnormal ocean environment data type set, no early warning is issued;

[0050] Through the above steps, the real-time marine environment data types are retrieved in the abnormal marine environment data types, which can timely identify and warn abnormal marine environment data.

[0051] A system for real-time monitoring and early warning of marine environment, comprising a data collection module, a marine environment type center data matrix module, a real-time marine environment data type module, and a module for judging whether the real-time marine environment data type is abnormal;

[0052] The data collection module is used to collect historical marine environmental data, extract features, construct a historical marine environmental data feature matrix, collect real-time marine environmental data through sensors and wireless network technology, and perform preprocessing;

[0053] The marine environment type central data matrix module classifies all marine environment data in the historical marine environment data feature matrix through a cluster analysis algorithm to obtain a marine environment type central data matrix, and sets an abnormal marine environment type data set through the marine environment type central data matrix;

[0054] The real-time ocean environment data type module obtains the real-time ocean environment data feature Euclidean distance set by calculating the distance between each feature data in the real-time ocean environment data feature set and the corresponding type center data in the ocean environment type center data matrix, and takes the type corresponding to the smallest Euclidean distance in the real-time ocean environment data feature Euclidean distance set as the type of the real-time ocean environment data to obtain the real-time ocean environment data type;

[0055] The module for determining whether the type of real-time ocean environment data is abnormal searches the type of real-time ocean environment data in a set of abnormal ocean environment data types. If the type of real-time ocean environment data belongs to a type in the set of abnormal ocean environment data types, an early warning is issued; if the type of real-time ocean environment data does not belong to a type in the set of abnormal ocean environment data types, no early warning is issued.

[0056] The present invention has the following beneficial effects:

[0057] The present invention proposes a method and system for real-time monitoring and early warning of the marine environment, which realizes real-time monitoring and early warning of the marine environment by collecting historical marine environmental data, constructing a feature matrix, using a cluster analysis algorithm for classification, and setting a set of abnormal marine environmental data types. The system uses sensors and wireless transmission technology to collect real-time marine environmental data, perform preprocessing, extract features, calculate Euclidean distance, and determine whether to issue an early warning. This system overcomes the problems of insufficient data processing capabilities, insufficient data analysis capabilities, and imperfect early warning response mechanisms in the prior art, and provides a scientific basis for marine environmental protection and disaster response.

[0058] Through cluster analysis algorithms and feature extraction, the system can efficiently process and analyze massive amounts of marine environmental data, improve data processing capabilities, and ensure the accuracy and reliability of data processing.

[0059] The system uses the abnormal ocean environment data type set constructed by historical data to more accurately determine whether real-time ocean environment data is abnormal, thereby improving the accuracy and reliability of early warnings.

[0060] Through real-time monitoring and early warning, the system can issue early warning information in a timely manner, guide relevant departments and personnel to take response measures, improve the early warning response mechanism, and improve the efficiency of responding to marine environmental issues.

[0061] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying any creative work.

[0063] Figure 1 The present invention is a flow chart of a method for real-time monitoring and early warning of a marine environment for determining whether to issue an early warning for a real-time marine environment by using a cluster analysis algorithm. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0065] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0066] Embodiment 1:

[0067] See also Figure 1 The present invention discloses a method for real-time monitoring and early warning of marine environment, comprising the following steps:

[0068] S1. Collect historical marine environment data, extract features, construct a feature matrix of historical marine environment data, use a cluster analysis algorithm to classify the marine environment data in the feature matrix of historical marine environment data, obtain a central data matrix of marine environment types, and set a set of abnormal marine environment data types according to the central data matrix of marine environment types;

[0069] The S1 comprises the following steps:

[0070] S11. Collect a large amount of historical marine environmental data, extract characteristic data (such as water temperature, salinity, dissolved oxygen, pH value), and construct the historical marine environmental data characteristic matrix A, as follows:

[0071]

[0072] Among them, A ib represents the bth feature of the i-th marine environmental data in the historical marine environmental data feature matrix, and a represents that there are a pieces of marine environmental data in the historical marine environmental data feature matrix;

[0073] S12, using a cluster analysis algorithm to classify the marine environment data of the historical marine environment data feature matrix to obtain a marine environment type center data matrix B, as follows:

[0074]

[0075] Among them, B ib represents the central data of all features b in the ith category in the central data matrix of marine environment categories, and c represents the total number of marine environment data categories in the central data matrix of marine environment categories;

[0076] The S12 comprises the following steps:

[0077] S121, randomly selecting several pieces of historical marine environment data from the historical marine environment data feature matrix as initial category centers;

[0078] S122, using the Euclidean distance formula to calculate the distance from each piece of historical marine environmental data in the historical marine environmental data feature matrix to each category center of the initial category center; the Euclidean distance formula is as follows:

[0079] Among them, p is any data in the historical marine environmental data feature matrix, q is initially the species center, and p i and q i represents the i-th feature of the p-th and q-th historical marine environmental data;

[0080] Assign each piece of historical marine environmental data in the historical marine environmental data feature matrix to the species where the species center closest to it is located;

[0081] Calculate the average value of all historical marine environmental data in each category as the new category center; the calculation formula of the average value is as follows:

[0082]

[0083] Among them, C j represents the new center of category j, D j represents the number of historical marine environmental data in category j, E ij represents the value of the jth feature of the i-th historical marine environmental data in category j;

[0084] S123, iterating step S122, setting the maximum number of iterations to e1, the current number of iterations to e2, and the minimum convergence threshold to f;

[0085] When the change of the species center is ≤f or the previous iteration number is e2 ≥ the maximum iteration number is e1, the iteration operation is stopped to obtain the marine environment species center data matrix;

[0086] S13, according to the marine environment type center data matrix B, set the abnormal marine environment data type set k = {k1, k2, ..., k i ,...,k l}, where k i represents the ith abnormal marine environment type in the abnormal marine environment data type set, l represents the total number of abnormal marine environment types in the abnormal marine environment data type set;

[0087] S2. Deploy sensors in the marine environment, transmit real-time marine environment data to a central monitoring system through wireless transmission technology, pre-process the real-time marine environment data, and obtain pre-processed real-time marine environment data;

[0088] The S2 comprises the following steps:

[0089] S21. Select appropriate sensor types, such as temperature sensors, salinity sensors, dissolved oxygen sensors, and pH sensors; determine the number, location, and distribution of sensors based on the characteristics and needs of the marine environment; install the sensors in the marine environment to ensure that they can work stably and collect accurate data;

[0090] S22. According to the characteristics and needs of the marine environment, select appropriate wireless transmission technologies, such as LoRa and 5G; establish a wireless transmission network to ensure that sensor data can be transmitted to the central monitoring system in real time to obtain real-time marine environment data;

[0091] S23, preprocessing the real-time ocean environment data to obtain preprocessed real-time ocean environment data;

[0092] The S23 comprises the following steps:

[0093] S231, for the missing data in the real-time marine environment data, fill them with interpolation and mean filling methods, identify and process outliers, delete outliers, and delete duplicate data records; obtain the cleaned real-time marine environment data;

[0094] S232, using a low-pass filter to remove high-frequency noise from the cleaned real-time ocean environment data, retaining low-frequency signals, decomposing the data using wavelet transform to remove noise, retaining useful information, and using linear regression or polynomial regression to remove trend items in the data to obtain denoised real-time ocean environment data;

[0095] S233, performing interpolation and normalization processing on the denoised real-time ocean environment data to obtain pre-processed real-time ocean environment data;

[0096] S3, extracting features from the pre-processed real-time marine environment data to construct a real-time marine environment data feature set, calculating the Euclidean distance from each feature data in the real-time marine environment data feature set to the corresponding type center data in the marine environment type center data matrix, obtaining a real-time marine environment data feature Euclidean distance set, taking the type corresponding to the smallest Euclidean distance in the real-time marine environment data feature Euclidean distance set as the type of the real-time marine environment data, and obtaining the real-time marine environment data type;

[0097] The S3 comprises the following steps:

[0098] S31, extracting features from the pre-processed real-time ocean environment data, and constructing a real-time ocean environment data feature set g = {g1, g2, ..., g i ,...,g b}, where g i represents the i-th feature of the real-time ocean environment data in the real-time ocean environment data feature set;

[0099] S32, the real-time ocean environment data feature set g={g1,g2,...,g i ,...,g b} to the corresponding type center data in the marine environment type center data matrix B, and obtain the real-time marine environment data feature Euclidean distance set h = {h1,h2,...,h i ,...,h c}, where h i Represents the Euclidean distance from the real-time marine environment data to the i-th species center data in the marine environment species center data matrix B;

[0100] S33, taking the real-time ocean environment data characteristic Euclidean distance set h={h1,h2,...,h i ,...,h c}, the type corresponding to the smallest Euclidean distance is taken as the type of real-time ocean environment data, and the type of real-time ocean environment data is obtained;

[0101] S4. Determine whether to issue an early warning according to the real-time marine environment type;

[0102] The S4 comprises the following steps:

[0103] S41, searching the type of real-time ocean environment data in the abnormal ocean environment data type set, if the type of real-time ocean environment data belongs to the type in the abnormal ocean environment data type set, issuing an early warning; if the type of real-time ocean environment data does not belong to the type in the abnormal ocean environment data type set, no early warning is issued.

[0104] Embodiment 2:

[0105] A marine environment real-time monitoring and early warning system, comprising a data collection module, a marine environment type center data matrix module, a real-time marine environment data type module, and a module for judging whether the real-time marine environment data type is abnormal;

[0106] The data collection module is used to collect historical marine environmental data, extract features, construct a historical marine environmental data feature matrix, collect real-time marine environmental data through sensors and wireless network technology, and perform preprocessing;

[0107] The marine environment type central data matrix module classifies all marine environment data in the historical marine environment data feature matrix through a cluster analysis algorithm to obtain a marine environment type central data matrix, and sets an abnormal marine environment type data set through the marine environment type central data matrix;

[0108] The real-time ocean environment data type module obtains the real-time ocean environment data feature Euclidean distance set by calculating the distance between each feature data in the real-time ocean environment data feature set and the corresponding type center data in the ocean environment type center data matrix, and takes the type corresponding to the smallest Euclidean distance in the real-time ocean environment data feature Euclidean distance set as the type of the real-time ocean environment data to obtain the real-time ocean environment data type;

[0109] The module for determining whether the type of real-time ocean environment data is abnormal searches the type of real-time ocean environment data in a set of abnormal ocean environment data types. If the type of real-time ocean environment data belongs to a type in the set of abnormal ocean environment data types, an early warning is issued; if the type of real-time ocean environment data does not belong to a type in the set of abnormal ocean environment data types, no early warning is issued.

[0110] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0111] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A method for real-time monitoring and early warning of marine environment, characterized in that: The following steps are involved: S1. Collect historical marine environment data, extract features, construct a feature matrix of historical marine environment data, use a cluster analysis algorithm to classify the marine environment data in the feature matrix of historical marine environment data, obtain a central data matrix of marine environment types, and set a set of abnormal marine environment data types according to the central data matrix of marine environment types; S2. Deploy sensors in the marine environment, transmit real-time marine environment data to a central monitoring system through wireless transmission technology, pre-process the real-time marine environment data, and obtain pre-processed real-time marine environment data; S3, extracting features from the pre-processed real-time marine environment data to construct a real-time marine environment data feature set, calculating the Euclidean distance from each feature data in the real-time marine environment data feature set to the corresponding type center data in the marine environment type center data matrix, obtaining a real-time marine environment data feature Euclidean distance set, taking the type corresponding to the smallest Euclidean distance in the real-time marine environment data feature Euclidean distance set as the type of the real-time marine environment data, and obtaining the real-time marine environment data type; S4. Determine whether to issue an early warning based on the real-time ocean environment type.

2. A method for real-time monitoring and early warning of marine environment according to claim 1, characterized in that: The S1 comprises the following steps: S11, collect a large amount of historical marine environmental data, extract feature data, and construct the historical marine environmental data feature matrix A, as follows: Among them, A ib represents the bth feature of the i-th marine environmental data in the historical marine environmental data feature matrix, and a represents that there are a pieces of marine environmental data in the historical marine environmental data feature matrix; S12, using a cluster analysis algorithm to classify the marine environment data of the historical marine environment data feature matrix to obtain a marine environment type center data matrix B, as follows: Among them, B ib represents the central data of all features b in the ith category in the central data matrix of marine environment categories, and c represents the total number of marine environment data categories in the central data matrix of marine environment categories; S13, according to the marine environment type center data matrix B, set the abnormal marine environment data type set k = {k1, k2, ..., k i ,...,k l }, where k i represents the ith abnormal marine environment type in the abnormal marine environment data type set, and l represents the total number of abnormal marine environment types in the abnormal marine environment data type set.

3. A method for real-time monitoring and early warning of marine environment according to claim 2, characterized in that: The S12 comprises the following steps: S121, randomly selecting several pieces of historical marine environment data from the historical marine environment data feature matrix as initial category centers; S122, using the Euclidean distance formula to calculate the distance from each piece of historical marine environmental data in the historical marine environmental data feature matrix to each category center of the initial category center; the Euclidean distance formula is as follows: Among them, p is any data in the historical marine environmental data feature matrix, q is initially the species center, and p i and q i represents the i-th feature of the p-th and q-th historical marine environmental data; Assign each piece of historical marine environmental data in the historical marine environmental data feature matrix to the species where the species center closest to it is located; Calculate the average value of all historical marine environmental data in each category as the new category center; the calculation formula of the average value is as follows: Among them, C j represents the new center of category j, D j represents the number of historical marine environmental data in category j, E ij represents the value of the jth feature of the i-th historical marine environmental data in category j; S123, iterating step S122, setting the maximum number of iterations to e1, the current number of iterations to e2, and the minimum convergence threshold to f; When the change of the species center is ≤f or the previous iteration number is e2 ≥ the maximum iteration number is e1, the iteration operation is stopped and the marine environment species center data matrix is ​​obtained.

4. A method for real-time monitoring and early warning of marine environment according to claim 1, characterized in that: The S2 comprises the following steps: S21. Select appropriate sensor types; determine the number, location and distribution of sensors according to the characteristics and requirements of the marine environment; install the sensors in the marine environment; S22. According to the characteristics and requirements of the marine environment, select appropriate wireless transmission technology; establish a wireless transmission network, and transmit real-time data to the central monitoring system through the wireless transmission network to obtain real-time marine environment data; S23, preprocessing the real-time ocean environment data to obtain preprocessed real-time ocean environment data.

5. A method for real-time monitoring and early warning of marine environment according to claim 4, characterized in that: The S23 comprises the following steps: S231, for the missing data in the real-time marine environment data, fill them with interpolation and mean filling methods, identify and process outliers, delete outliers, and delete duplicate data records; obtain the cleaned real-time marine environment data; S232, using a low-pass filter to remove high-frequency noise from the cleaned real-time ocean environment data, retaining low-frequency signals, decomposing the data using wavelet transform to remove noise, retaining useful information, and using linear regression or polynomial regression to remove trend items in the data to obtain denoised real-time ocean environment data; S233, interpolating and normalizing the denoised real-time ocean environment data to obtain pre-processed real-time ocean environment data.

6. A method for real-time monitoring and early warning of marine environment according to claim 1, characterized in that: The S3 comprises the following steps: S31, extracting features from the pre-processed real-time ocean environment data, and constructing a real-time ocean environment data feature set g = {g1, g2, ..., g i ,...,g b }, where g i represents the i-th feature of the real-time ocean environment data in the real-time ocean environment data feature set; S32, the real-time ocean environment data feature set g={g1,g2,...,g i ,...,g b } to the corresponding type center data in the marine environment type center data matrix B, and obtain the real-time marine environment data feature Euclidean distance set h = {h1,h2,...,h i ,...,h c }, where h i Represents the Euclidean distance from the real-time marine environment data to the i-th species center data in the marine environment species center data matrix B; S33, taking the real-time ocean environment data characteristic Euclidean distance set h={h1,h2,...,h i ,...,h c The type corresponding to the smallest Euclidean distance in} is taken as the type of real-time ocean environment data, and the real-time ocean environment data type is obtained.

7. A method for real-time monitoring and early warning of marine environment according to claim 1, characterized in that: The S4 comprises the following steps: S41, searching the type of real-time ocean environment data in the abnormal ocean environment data type set, if the type of real-time ocean environment data belongs to the type in the abnormal ocean environment data type set, issuing an early warning; if the type of real-time ocean environment data does not belong to the type in the abnormal ocean environment data type set, no early warning is issued.

8. A system for implementing a marine environment real-time monitoring and early warning method as claimed in any one of claims 1 to 7, characterized in that: Data collection module, marine environment type center data matrix module, real-time marine environment data type module, and module for judging whether the real-time marine environment data type is abnormal; The data collection module is used to collect historical marine environmental data, extract features, construct a historical marine environmental data feature matrix, collect real-time marine environmental data through sensors and wireless network technology, and perform preprocessing; The marine environment type central data matrix module classifies all marine environment data in the historical marine environment data feature matrix through a cluster analysis algorithm to obtain a marine environment type central data matrix, and sets an abnormal marine environment type data set through the marine environment type central data matrix; The real-time ocean environment data type module obtains the real-time ocean environment data feature Euclidean distance set by calculating the distance between each feature data in the real-time ocean environment data feature set and the corresponding type center data in the ocean environment type center data matrix, and takes the type corresponding to the smallest Euclidean distance in the real-time ocean environment data feature Euclidean distance set as the type of the real-time ocean environment data to obtain the real-time ocean environment data type; The module for determining whether the type of real-time ocean environment data is abnormal searches the type of real-time ocean environment data in a set of abnormal ocean environment data types. If the type of real-time ocean environment data belongs to a type in the set of abnormal ocean environment data types, an early warning is issued; if the type of real-time ocean environment data does not belong to a type in the set of abnormal ocean environment data types, no early warning is issued.

Images