Marine emergency analysis and response decision system

Abstract

The present invention discloses a marine emergency analysis and response decision system, which relates to the technical field of marine ecological data analysis, including: a data acquisition module for collecting time series data of influencing factors of red tide organisms in a monitoring area; a basic risk assessment module for assessing the basic risk index of red tide organism outbreaks in each monitoring area; a risk coefficient assessment module for assessing the risk coefficient of red tide organism outbreaks in each monitoring area; a state correction module for correcting the basic risk index of red tide organism outbreaks in the monitoring area using the risk coefficient of red tide organism outbreaks in the monitoring area, and obtaining the risk index of red tide organism outbreaks in each monitoring area; a monitoring module for judging whether the red tide outbreak risk index in each monitoring area exceeds the tolerable threshold of red tide in the corresponding monitoring area, and if so, issuing an alarm. The advantages of the present invention are: it can warn of red tide risks in advance and reduce the economic losses and ecological damage caused by red tide disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of marine ecological data analysis, and in particular to a marine emergency analysis and response decision-making system. Background Art

[0002] Marine emergency analysis refers to a comprehensive and in-depth analysis of emergency events that may occur or have occurred in the marine environment to assess their impact, develop response strategies and guide emergency response actions. Its purpose is to improve the early warning capability of marine disasters, emergency response efficiency and disaster recovery capabilities, thereby minimizing the losses caused by disasters.

[0003] Existing marine red tide emergency analysis mainly relies on field experts to conduct objective analysis of water quality data in the current area, but red tide biological outbreaks are usually multi-factorial and easily affected by the natural environment, making the results obtained by conventional empirical analysis inaccurate. Summary of the invention

[0004] In order to solve the above technical problems, a marine emergency analysis and response decision-making system is provided. This technical solution solves the problem that the existing marine red tide emergency analysis mainly relies on field experts to objectively analyze the water quality data of the current area, but the red tide biological outbreak is usually multi-factorial and easily affected by the natural environment, making the results obtained by conventional empirical analysis inaccurate.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] Marine emergency analysis and response decision-making system, including:

[0007] Data acquisition module, which is used to obtain real-time ocean water quality data in all monitoring areas, and select the intrinsic water quality parameters that meet the preferences of red tide organisms in the real-time ocean water quality data per unit time according to the preferences of red tide organisms, and obtain the time series data of the influencing factors of red tide organisms in several monitoring areas;

[0008] A basic risk assessment module, the basic risk assessment module is electrically connected to the data acquisition module, and the basic risk assessment module is used to analyze the development trend of red tide organisms in each monitoring area based on the time series data of the influencing factors of red tide organisms in several monitoring areas, and to assess the basic risk index of red tide organism outbreaks in each monitoring area;

[0009] Risk factor assessment module: The risk factor assessment module is used to obtain real-time marine natural environment data in all monitoring areas, analyze the marine environment status in each monitoring area, and assess the risk factor of red tide biological outbreaks in each monitoring area;

[0010] A state correction module, the state correction module is electrically connected to the basic risk assessment module and the risk coefficient assessment module, and the state correction module is used to use the red tide biological outbreak risk coefficient in each monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area, and obtain the red tide biological outbreak risk index in each monitoring area;

[0011] The monitoring module is electrically connected to the state correction module. The monitoring module is used to determine the red tide tolerable threshold in each monitoring area based on the marine aquaculture species in all monitoring areas, and to judge whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerable threshold in the corresponding monitoring area. If not, continue to observe; if so, issue an alarm.

[0012] Preferably, the data acquisition module internally includes:

[0013] The preference characteristic unit determines the preference characteristic data of each red tide organism based on several red tide organisms in all monitoring areas, and constructs a preference characteristic data array of the red tide organisms;

[0014] The preference model unit, based on the preference characteristic data array of red tide organisms, takes the preference characteristic data of each red tide organism as the leaf node of the decision tree, constructs a preference recognition model for each red tide organism, and forms a random forest for preference recognition of red tide organisms;

[0015] The data partitioning unit, based on the random forest for identifying the preference of red tide organisms, takes the real-time ocean water quality data per unit time as input, calculates the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the partitioning condition, and performs several rounds of iterations to obtain the time series data of the influencing factors of red tide organisms in several monitoring areas;

[0016] The calculation of the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the segmentation condition is specifically as follows:

[0017] ;

[0018] In the formula, Preference characteristic data for red tide organisms For real-time ocean water quality data The information gain value of Real-time ocean water quality data The entropy value of According to the red tide organisms The subsets after the real-time ocean water quality data is divided by preference feature data, An array of data on the preference characteristics of red tide organisms.

[0019] Preferably, the basic risk assessment module includes:

[0020] A preprocessing unit performs data preprocessing based on the time series data of factors affecting red tide organisms in several monitoring areas;

[0021] The correlation coefficient unit is based on the time series data of the influencing factors of red tide organisms in several monitoring areas. The linear regression coefficient between the preference characteristic data of red tide organisms and the time series data of the influencing factors in each monitoring area under unit time is calculated and recorded as the correlation coefficient of the influencing factors of red tide organisms.

[0022] A weighting unit assigns weights to the time series data of the influencing factors of red tide organisms in several monitoring areas according to the correlation coefficients of the influencing factors of red tide organisms, so as to obtain the time series weighted data of the influencing factors of red tide organisms in several monitoring areas;

[0023] The risk value calculation unit calculates the increased risk value of the outbreak of red tide organisms in the monitoring area per unit time based on the time series weighted data of the influencing factors of red tide organisms in the monitoring areas and the time series data of the influencing factors of red tide organisms in the monitoring areas;

[0024] Red tide model unit, to establish a red tide outbreak analysis model within the monitoring area;

[0025] The red tide outbreak analysis unit substitutes the time series data of the influencing factors of red tide organisms in several monitoring areas into the red tide outbreak analysis model in the monitoring area corresponding to the increased risk value of red tide organism outbreak in the monitoring area per unit time, and analyzes the basic risk index of red tide organism outbreak in each monitoring area per unit time;

[0026] Among them, the correlation coefficient between the red tide organisms and the internal influencing factors is specifically:

[0027] ;

[0028] in, is the correlation coefficient of the factors affecting red tide organisms, In the monitoring area The first Preference feature data, is the first unit time in the monitoring area Time series data of factors affecting red tide organisms, n is the total number of red tide organisms in the monitoring area;

[0029] The increased risk value of red tide outbreak in the monitoring area under the calculation unit time is specifically:

[0030] ;

[0031] In the formula, is the jth increased risk value of red tide biological outbreak in the tth unit time in the monitoring area, is the first unit time in the monitoring area The k-th influencing factor time series weighted data of red tide organisms, is the first unit time in the monitoring area The time series data of the kth influencing factor of red tide organisms;

[0032] The red tide outbreak analysis model in the monitoring area is specifically as follows:

[0033] ;

[0034] In the formula, For the In the monitoring area The first The basic risk index of red tide biological outbreaks is For the In the monitoring area The first Increased risk of red tide outbreaks For the In the monitoring area The unit time The risk of red tide outbreaks increases by an error value. is the linear regression coefficient, is the moving average coefficient, is the total number of autoregressive terms, is the total number of moving average items.

[0035] Preferably, the risk factor assessment module includes:

[0036] The historical natural data unit obtains the marine natural environment data of red tide biological outbreaks in several historical monitoring areas and constructs a red tide biological outbreak control data array; the marine natural environment data includes: topographic characteristics, hydrological characteristics, meteorological characteristics, and ocean current characteristics;

[0037] The matching verification unit recursively calculates the matching degree of the marine natural environment of the red tide biological outbreak in each monitoring area based on the real-time marine natural environment data and the red tide biological outbreak control data array in all monitoring areas;

[0038] The risk coefficient unit normalizes the matching degree of the red tide biological outbreak marine natural environment in each monitoring area to obtain the red tide biological outbreak risk coefficient in each monitoring area;

[0039] Among them, the red tide biological outbreak risk coefficient in each monitoring area is as follows:

[0040] ;

[0041] In the formula, For the In the monitoring area The risk factor of red tide biological outbreak per unit time is For the In the monitoring area The matching degree of red tide organisms to the natural environment per unit time, For the In the monitoring area The first Real-time marine natural environment data, For the history In the monitoring area Marine natural environment data, is the total number of monitoring areas.

[0042] Preferably, the state correction module internally includes:

[0043] A risk status acquisition unit is used to acquire the basic risk index and risk coefficient of red tide organism outbreak in several monitoring areas;

[0044] The state correction unit uses each monitoring area as a unique key value, and uses the red tide biological outbreak risk coefficient in the monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area to obtain the red tide biological outbreak risk index in each monitoring area;

[0045] The specific red tide biological outbreak risk indicators in the monitoring area are:

[0046] ;

[0047] In the formula, For the In the monitoring area The red tide biological outbreak risk index per unit time.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention proposes a marine emergency analysis and response decision-making plan, which screens out key factors that promote red tide outbreaks from water quality data, analyzes the basic risk index of red tide biological outbreaks, combines real-time marine natural environmental conditions, evaluates the impact of the environment on red tide outbreaks, and calculates a more accurate red tide outbreak risk coefficient to correct the basic risk index of red tide biological outbreaks, and obtains the red tide outbreak risk index in each monitoring area; sets a threshold value according to the red tide tolerance capacity of the aquaculture species in each monitoring area, and determines whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerance threshold in the corresponding monitoring area, and decides whether to issue an alarm. The advantages of the present invention are: it can warn of red tide risks in advance and reduce the economic losses and ecological damage caused by red tide disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is the framework diagram of the marine emergency analysis and response decision system;

[0051] Figure 2 This is the internal framework diagram of the data acquisition module;

[0052] Figure 3 This is the internal framework diagram of the basic risk assessment module;

[0053] Figure 4 This is the internal framework diagram of the risk factor assessment module;

[0054] Figure 5 This is the internal framework diagram of the status correction module. DETAILED DESCRIPTION

[0055] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0056] Reference Figure 1 As shown, the marine emergency analysis and response decision-making system includes:

[0057] Data acquisition module, which is used to obtain real-time ocean water quality data in all monitoring areas, and select the intrinsic water quality parameters that meet the preferences of red tide organisms in the real-time ocean water quality data per unit time according to the preferences of red tide organisms, and obtain the time series data of the influencing factors of red tide organisms in several monitoring areas;

[0058] A basic risk assessment module, the basic risk assessment module is electrically connected to the data acquisition module, and the basic risk assessment module is used to analyze the development trend of red tide organisms in each monitoring area based on the time series data of the influencing factors of red tide organisms in several monitoring areas, and to assess the basic risk index of red tide organism outbreaks in each monitoring area;

[0059] Risk factor assessment module: The risk factor assessment module is used to obtain real-time marine natural environment data in all monitoring areas, analyze the marine environment status in each monitoring area, and assess the risk factor of red tide biological outbreaks in each monitoring area;

[0060] A state correction module, the state correction module is electrically connected to the basic risk assessment module and the risk coefficient assessment module, and the state correction module is used to use the red tide biological outbreak risk coefficient in each monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area, and obtain the red tide biological outbreak risk index in each monitoring area;

[0061] The monitoring module is electrically connected to the state correction module. The monitoring module is used to determine the red tide tolerable threshold in each monitoring area based on the marine aquaculture species in all monitoring areas, and to judge whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerable threshold in the corresponding monitoring area. If not, continue to observe; if so, issue an alarm.

[0062] It should be noted that due to the different types of aquatic product farming in the monitoring area, the red tide tolerance threshold in the corresponding monitoring area needs to determine an actual threshold range based on the actual farming species, and this part of the content is well known to practitioners in this field and will not be described in detail here.

[0063] This solution screens out the key factors that promote the outbreak of red tides from water quality data, analyzes the basic risk index of red tide biological outbreaks, combines real-time marine natural environmental conditions, evaluates the impact of the environment on red tide outbreaks, and calculates a more accurate red tide outbreak risk coefficient to correct the basic risk index of red tide biological outbreaks, and obtains the red tide outbreak risk index in each monitoring area; sets a threshold value according to the red tide tolerance of the aquaculture species in each monitoring area, and determines whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerance threshold in the corresponding monitoring area, and decides whether to issue an alarm. The advantages of the present invention are: it can warn of red tide risks in advance and reduce the economic losses and ecological damage caused by red tide disasters.

[0064] Reference Figure 2 As shown, the data acquisition module includes:

[0065] The preference characteristic unit determines the preference characteristic data of each red tide organism based on several red tide organisms in all monitoring areas, and constructs a preference characteristic data array of the red tide organisms;

[0066] The preference model unit, based on the preference characteristic data array of red tide organisms, takes the preference characteristic data of each red tide organism as the leaf node of the decision tree, constructs a preference recognition model for each red tide organism, and forms a random forest for preference recognition of red tide organisms;

[0067] The data partitioning unit, based on the random forest for identifying the preference of red tide organisms, takes the real-time ocean water quality data per unit time as input, calculates the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the partitioning condition, and performs several rounds of iterations to obtain the time series data of the influencing factors of red tide organisms in several monitoring areas;

[0068] The calculation of the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the segmentation condition is specifically as follows:

[0069] ;

[0070] In the formula, Preference characteristic data for red tide organisms For real-time ocean water quality data The information gain value of Real-time ocean water quality data The entropy value of According to the red tide organisms The subsets after the real-time ocean water quality data is divided by preference feature data, It is the data array of the preference characteristics of red tide organisms.

[0071] It is understandable that since red tide organisms usually include many types, and different red tide organisms have different preferences for substances in water quality, only analyzing substances in water quality will lead to a large difference in the final prediction results. Therefore, by analyzing the types of red tide organisms in all monitoring areas, corresponding data division is carried out to improve the accuracy of the data;

[0072] For example, red tide organisms in a certain monitoring area include foraminifera, dinoflagellates, and dinoflagellates. The preferences of the corresponding red tide organisms are as follows: foraminifera prefer waters with high nutrient concentrations; dinoflagellates prefer warm and well-lit environments; and dinoflagellates are sensitive to salinity and temperature. Based on this, data is divided according to the preference characteristic data of red tide organisms for real-time ocean water quality data to improve the accuracy of subsequent model construction and prediction.

[0073] Reference Figure 3 As shown, the basic risk assessment module includes:

[0074] A preprocessing unit performs data preprocessing based on the time series data of factors affecting red tide organisms in several monitoring areas;

[0075] The correlation coefficient unit is based on the time series data of the influencing factors of red tide organisms in several monitoring areas. The linear regression coefficient between the preference characteristic data of red tide organisms and the time series data of the influencing factors in each monitoring area under unit time is calculated and recorded as the correlation coefficient of the influencing factors of red tide organisms.

[0076] A weighting unit assigns weights to the time series data of the influencing factors of red tide organisms in several monitoring areas according to the correlation coefficients of the influencing factors of red tide organisms, so as to obtain the time series weighted data of the influencing factors of red tide organisms in several monitoring areas;

[0077] The risk value calculation unit calculates the increased risk value of the outbreak of red tide organisms in the monitoring area per unit time based on the time series weighted data of the influencing factors of red tide organisms in the monitoring areas and the time series data of the influencing factors of red tide organisms in the monitoring areas;

[0078] Red tide model unit, to establish a red tide outbreak analysis model within the monitoring area;

[0079] The red tide outbreak analysis unit substitutes the time series data of the influencing factors of red tide organisms in several monitoring areas into the red tide outbreak analysis model in the monitoring area corresponding to the increased risk value of red tide organism outbreak in the monitoring area per unit time, and analyzes the basic risk index of red tide organism outbreak in each monitoring area per unit time;

[0080] Among them, the correlation coefficient between the red tide organisms and the internal influencing factors is specifically:

[0081] ;

[0082] in, is the correlation coefficient of the factors affecting red tide organisms, In the monitoring area The first Preference feature data, is the first unit time in the monitoring area Time series data of factors affecting red tide organisms, n is the total number of red tide organisms in the monitoring area;

[0083] The increased risk value of red tide outbreak in the monitoring area under the calculation unit time is specifically:

[0084] ;

[0085] In the formula, is the jth increased risk value of red tide biological outbreak in the tth unit time in the monitoring area, is the first unit time in the monitoring area The k-th influencing factor time series weighted data of red tide organisms, is the first unit time in the monitoring area The time series data of the kth influencing factor of red tide organisms;

[0086] The red tide outbreak analysis model in the monitoring area is specifically as follows:

[0087] ;

[0088] In the formula, For the In the monitoring area The unit time The basic risk index of red tide biological outbreaks is For the In the monitoring area The unit time Increased risk of red tide outbreaks For the In the monitoring area The unit time The risk of red tide outbreaks increases by an error value. is the linear regression coefficient, is the moving average coefficient, is the total number of autoregressive terms, is the total number of moving average items.

[0089] It can be understood that the larger the correlation coefficient of the influencing factors of red tide organisms, the greater the impact of the ocean water quality in the monitoring area at the current time on the increase of red tide organisms, and the smaller the correlation coefficient of the influencing factors of red tide organisms, the smaller the impact of the ocean water quality in the monitoring area at the current time on the increase of red tide organisms;

[0090] This program integrates and analyzes the time series data of factors affecting red tide organisms in multiple monitoring areas, uses linear regression and weighted processing to accurately quantify the contribution of each factor to the outbreak of red tide organisms, and constructs a risk analysis model to evaluate the basic risk index of red tide organism outbreaks in each region in real time, thereby improving the accuracy of early warning and the efficiency of prevention and control.

[0091] Reference Figure 4 As shown in the figure, the risk factor assessment module includes:

[0092] The historical natural data unit obtains the marine natural environment data of red tide biological outbreaks in several historical monitoring areas and constructs a red tide biological outbreak control data array; the marine natural environment data includes: topographic characteristics, hydrological characteristics, meteorological characteristics, and ocean current characteristics;

[0093] The matching verification unit recursively calculates the matching degree of the marine natural environment of the red tide biological outbreak in each monitoring area based on the real-time marine natural environment data and the red tide biological outbreak control data array in all monitoring areas;

[0094] The risk coefficient unit normalizes the matching degree of the red tide biological outbreak marine natural environment in each monitoring area to obtain the red tide biological outbreak risk coefficient in each monitoring area;

[0095] Among them, the red tide biological outbreak risk coefficient in each monitoring area is as follows:

[0096] ;

[0097] In the formula, For the In the monitoring area The risk factor of red tide biological outbreak per unit time is For the In the monitoring area The matching degree of red tide organisms to the natural environment per unit time, For the In the monitoring area The unit time Real-time marine natural environment data, For the history In the monitoring area Marine natural environment data, is the total number of monitoring areas.

[0098] It should be noted that due to the non-intervention of marine environmental data and the fact that the acquired data do not conform to the normal distribution, the data correlation verification effect at a fine-grained level is poor. The overall data dimension vector distance matching is performed at a coarse-grained level at high latitudes to evaluate the distance matching degree between the current marine environmental data and the marine environmental data before the historical red tide outbreak, so as to improve the accuracy of the red tide biological outbreak risk coefficient in each monitoring area.

[0099] Reference Figure 5 As shown, the state correction module includes:

[0100] A risk status acquisition unit is used to acquire the basic risk index and risk coefficient of red tide organism outbreak in several monitoring areas;

[0101] The state correction unit uses each monitoring area as a unique key value, and uses the red tide biological outbreak risk coefficient in the monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area to obtain the red tide biological outbreak risk index in each monitoring area;

[0102] The specific red tide biological outbreak risk indicators in the monitoring area are:

[0103] ;

[0104] In the formula, For the In the monitoring area The red tide biological outbreak risk index per unit time.

[0105] The use process of the present invention is:

[0106] Step 1: Based on several red tide organisms in all monitoring areas, determine the preference characteristic data of each red tide organism and construct the preference characteristic data array of the red tide organisms;

[0107] Step 2: Based on the preference feature data array of red tide organisms, the preference feature data of each red tide organism is used as the leaf node of the decision tree, a preference recognition model for each red tide organism is constructed, and a random forest for preference recognition of red tide organisms is formed;

[0108] Step 3: Based on the random forest for identifying the preference of red tide organisms, the real-time ocean water quality data per unit time is used as input, and the maximum value of the information gain between the real-time ocean water quality data and each leaf node is calculated as the segmentation condition for several rounds of iterations to obtain the time series data of the influencing factors of red tide organisms in several monitoring areas;

[0109] Step 4: Data preprocessing based on the time series data of factors affecting red tide organisms in several monitoring areas;

[0110] Step 5: Based on the time series data of the influencing factors of red tide organisms in several monitoring areas, the linear regression coefficient between the preference characteristic data of red tide organisms and the time series data of the influencing factors in each monitoring area per unit time is calculated, which is recorded as the correlation coefficient of the influencing factors of red tide organisms;

[0111] Step 6: assign weights to the time series data of the influencing factors of red tide organisms in several monitoring areas according to the correlation coefficients of the influencing factors of red tide organisms, and obtain the time series weighted data of the influencing factors of red tide organisms in several monitoring areas;

[0112] Step 7: Based on the time series weighted data of the influencing factors of red tide organisms in several monitoring areas and the time series data of the influencing factors of red tide organisms in several monitoring areas, calculate the increased value of the red tide organism outbreak risk in the monitoring area per unit time;

[0113] Step 8: Establish a red tide outbreak analysis model in the monitoring area;

[0114] Step 9: Substitute the increased risk value of red tide organism outbreak in the monitoring area corresponding to the time series data of the influencing factors of red tide organisms in several monitoring areas per unit time into the red tide outbreak analysis model in the monitoring area, and analyze the basic risk index of red tide organism outbreak in each monitoring area per unit time;

[0115] Step 10: Obtain the marine natural environment data of red tide organisms outbreaks in several historical monitoring areas, and construct a red tide organisms outbreak control data array;

[0116] Step 11: Based on the real-time marine natural environment data and the red tide biological outbreak control data array in all monitoring areas, recursively calculate the matching degree of the red tide biological outbreak marine natural environment in each monitoring area;

[0117] Step 12: Normalize the matching degree of the red tide organism outbreak marine natural environment in each monitoring area to obtain the red tide organism outbreak risk coefficient in each monitoring area;

[0118] Step 13: Obtain the red tide organism outbreak basic risk index and red tide organism outbreak risk coefficient in several monitoring areas;

[0119] Step 14: Taking each monitoring area as a unique key value, using the red tide biological outbreak risk coefficient in the monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area, and obtaining the red tide biological outbreak risk index in each monitoring area;

[0120] Step 15: Based on the marine aquaculture species in all monitoring areas, determine the red tide tolerance threshold in each monitoring area, and judge whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerance threshold in the corresponding monitoring area. If not, continue to observe; if so, issue an alarm.

[0121] In summary, the advantages of the present invention are: by screening out the key factors that promote the outbreak of red tides from water quality data, analyzing the basic risk index of the outbreak of red tide organisms, combining the real-time marine natural environment conditions, evaluating the impact of the environment on the outbreak of red tides, and calculating a more accurate red tide outbreak risk coefficient to correct the basic risk index of the outbreak of red tide organisms, and obtaining the red tide outbreak risk index in each monitoring area; setting a threshold value according to the red tide tolerance of the aquaculture species in each monitoring area, and judging whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerance threshold in the corresponding monitoring area, and deciding whether to issue an alarm. The advantages of the present invention are: being able to warn of red tide risks in advance and reduce the economic losses and ecological damage caused by red tide disasters.

[0122] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. Marine emergency analysis and response decision-making system, characterized by: include: Data acquisition module, which is used to obtain real-time ocean water quality data in all monitoring areas, and select the intrinsic water quality parameters that meet the preferences of red tide organisms in the real-time ocean water quality data per unit time according to the preferences of red tide organisms, and obtain the time series data of the influencing factors of red tide organisms in several monitoring areas; A basic risk assessment module, the basic risk assessment module is electrically connected to the data acquisition module, and the basic risk assessment module is used to analyze the development trend of red tide organisms in each monitoring area based on the time series data of the influencing factors of red tide organisms in several monitoring areas, and to assess the basic risk index of red tide organism outbreaks in each monitoring area; Risk factor assessment module: The risk factor assessment module is used to obtain real-time marine natural environment data in all monitoring areas, analyze the marine environment status in each monitoring area, and assess the risk factor of red tide biological outbreaks in each monitoring area; A state correction module, the state correction module is electrically connected to the basic risk assessment module and the risk coefficient assessment module, and the state correction module is used to use the red tide biological outbreak risk coefficient in each monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area, and obtain the red tide biological outbreak risk index in each monitoring area; The monitoring module is electrically connected to the state correction module. The monitoring module is used to determine the red tide tolerance threshold in each monitoring area based on the marine aquaculture species in all monitoring areas, and to judge whether the red tide outbreak risk index in each monitoring area exceeds the red tide tolerance threshold in the corresponding monitoring area. If not, continue to observe, and if so, issue an alarm; The basic risk assessment module includes: A preprocessing unit performs data preprocessing based on the time series data of factors affecting red tide organisms in several monitoring areas; The correlation coefficient unit is based on the time series data of the influencing factors of red tide organisms in several monitoring areas. The linear regression coefficient between the preference characteristic data of red tide organisms and the time series data of the influencing factors in each monitoring area under unit time is calculated and recorded as the correlation coefficient of the influencing factors of red tide organisms. A weighting unit assigns weights to the time series data of the influencing factors of red tide organisms in several monitoring areas according to the correlation coefficients of the influencing factors of red tide organisms, so as to obtain the time series weighted data of the influencing factors of red tide organisms in several monitoring areas; The risk value calculation unit calculates the increased risk value of the outbreak of red tide organisms in the monitoring area per unit time based on the time series weighted data of the influencing factors of red tide organisms in the monitoring areas and the time series data of the influencing factors of red tide organisms in the monitoring areas; Red tide model unit, to establish a red tide outbreak analysis model within the monitoring area; The red tide outbreak analysis unit substitutes the time series data of the influencing factors of red tide organisms in several monitoring areas into the red tide outbreak analysis model in the monitoring area corresponding to the increased risk value of red tide organism outbreak in the monitoring area per unit time, and analyzes the basic risk index of red tide organism outbreak in each monitoring area per unit time; Among them, the correlation coefficient between the red tide organisms and the internal influencing factors is specifically: ； in, is the correlation coefficient of the factors affecting red tide organisms, In the monitoring area The first Preference feature data, is the first unit time in the monitoring area Time series data of factors affecting red tide organisms, n is the total number of red tide organisms in the monitoring area; The increased risk value of red tide outbreak in the monitoring area under the calculation unit time is specifically: ； In the formula, is the jth increased risk value of red tide biological outbreak in the tth unit time in the monitoring area, is the first unit time in the monitoring area The k-th influencing factor time series weighted data of red tide organisms, is the first unit time in the monitoring area The time series data of the kth influencing factor of red tide organisms; The red tide outbreak analysis model in the monitoring area is specifically as follows: ； In the formula, For the In the monitoring area The unit time The basic risk index of red tide biological outbreaks is For the In the monitoring area The unit time Increased risk of red tide outbreaks For the In the monitoring area The unit time The risk of red tide outbreaks increases by an error value. is the linear regression coefficient, is the moving average coefficient, is the total number of autoregressive terms, is the total number of moving average items; The risk factor assessment module includes: The historical natural data unit obtains the marine natural environment data of red tide biological outbreaks in several historical monitoring areas and constructs a red tide biological outbreak control data array; the marine natural environment data includes: topographic characteristics, hydrological characteristics, meteorological characteristics, and ocean current characteristics; The matching verification unit recursively calculates the matching degree of the marine natural environment of the red tide biological outbreak in each monitoring area based on the real-time marine natural environment data and the red tide biological outbreak control data array in all monitoring areas; The risk coefficient unit normalizes the matching degree of the red tide biological outbreak marine natural environment in each monitoring area to obtain the red tide biological outbreak risk coefficient in each monitoring area; The specific risk factor of red tide biological outbreak in each monitoring area is: ； In the formula, For the In the monitoring area The risk factor of red tide biological outbreak per unit time is For the In the monitoring area The matching degree of red tide organisms to the natural environment per unit time, For the In the monitoring area The unit time Real-time marine natural environment data, For the history In the monitoring area Marine natural environment data, is the total number of monitoring areas; Among them, the state correction module includes: A risk status acquisition unit is used to acquire the basic risk index and risk coefficient of red tide organism outbreak in several monitoring areas; The state correction unit uses each monitoring area as a unique key value, and uses the red tide biological outbreak risk coefficient in the monitoring area to correct the red tide biological outbreak basic risk index in each monitoring area to obtain the red tide biological outbreak risk index in each monitoring area; The specific red tide biological outbreak risk indicators in the monitoring area are: ； In the formula, For the In the monitoring area The red tide biological outbreak risk index per unit time.

2. The marine emergency analysis and response decision system according to claim 1 is characterized in that: The data acquisition module includes: The preference characteristic unit determines the preference characteristic data of each red tide organism based on several red tide organisms in all monitoring areas, and constructs a preference characteristic data array of the red tide organisms; The preference model unit, based on the preference characteristic data array of red tide organisms, takes the preference characteristic data of each red tide organism as the leaf node of the decision tree, constructs a preference recognition model for each red tide organism, and forms a random forest for preference recognition of red tide organisms; The data partitioning unit, based on the random forest for identifying the preference of red tide organisms, takes the real-time ocean water quality data per unit time as input, calculates the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the partitioning condition, and performs several rounds of iterations to obtain the time series data of the influencing factors of red tide organisms in several monitoring areas; The calculation of the maximum value of the information gain between the real-time ocean water quality data and each leaf node as the segmentation condition is specifically as follows: ； In the formula, Preference characteristic data for red tide organisms For real-time ocean water quality data The information gain value of Real-time ocean water quality data The entropy value of According to the red tide organisms The subsets after the real-time ocean water quality data is divided by preference feature data, An array of data on the preference characteristics of red tide organisms.

Images