Marine ecological environment assessment early warning system based on big data and AI

Abstract

The invention relates to the technical field of marine ecological environment monitoring, in particular to a marine ecological environment assessment early warning system based on big data and AI, which comprises a multi-source data fusion module, a dynamic feature extraction module, a meteorological disturbance coefficient calculation module, an ecological index generation module and a coupling early warning generation module. Wherein the multi-source data fusion module is used for collecting multi-source data in real time; the dynamic feature extraction module is used for extracting 72-hour change gradients of the three types of core parameters; the meteorological disturbance coefficient calculation module is used for calculating and outputting a meteorological disturbance coefficient; the ecological index generation module is used for generating a water quality deterioration index and a biological stress index; and the coupling early warning generation module is used for outputting an ecological risk level signal. According to the method, meteorological interference is reduced in real time, water quality deterioration and biological stress indexes are subjected to weighted fusion, the marine ecological environment risk is accurately evaluated, and scientific support is provided for ecological protection and pollution early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of marine ecological environment monitoring, and particularly relates to a marine ecological environment evaluation and early warning system based on big data and AI. BACKGROUND

[0002] Marine water pollution and biological stress have become a global environmental problem that needs to be solved urgently. Key ecological parameters such as dissolved oxygen concentration, chlorophyll concentration and plankton biomass are the core indicators for measuring the health of marine water bodies. At the same time, meteorological factors such as wind speed, precipitation and air temperature also have an important impact on the marine ecological environment.

[0003] Although the existing marine ecological evaluation method can analyze some basic ecological indicators, it lacks effective real-time early warning and risk assessment capabilities when considering the complex influence of meteorological factors on water quality and biological communities. Therefore, there is an urgent need for a marine ecological environment evaluation and early warning system based on big data and AI to solve the above problems and achieve accurate real-time evaluation of the marine ecological environment. SUMMARY

[0004] Based on the above purpose, the present application provides a marine ecological environment evaluation and early warning system based on big data and AI.

[0005] A marine ecological environment evaluation and early warning system based on big data and AI, comprising a multi-source data fusion module, a dynamic feature extraction module, a meteorological disturbance coefficient calculation module, an ecological index generation module and a coupling early warning generation module; wherein: The multi-source data fusion module is used to collect satellite remote sensing data, marine sensor network data and meteorological station data in real time, and generate standardized ecological data streams through spatio-temporal alignment processing; The dynamic feature extraction module is used to receive the standardized ecological data stream and extract the 72-hour change gradient of the three core parameters of dissolved oxygen concentration, chlorophyll concentration and plankton biomass using a sliding window mechanism; The meteorological disturbance coefficient calculation module is based on a meteorological disturbance sensitivity matrix established from a historical disaster database and calculates and outputs the meteorological disturbance coefficient; The ecological index generation module is used to receive the 72-hour change gradient of the three core parameters, generate a water quality deterioration index according to the negative correlation intensity of dissolved oxygen concentration and chlorophyll concentration, and generate a biological stress index according to the positive correlation intensity of chlorophyll concentration and plankton biomass; The coupling early warning generation module is used to proportionally reduce the meteorological interference components in the water quality deterioration index and the biological stress index when the meteorological disturbance coefficient exceeds the preset threshold, and to output the ecological risk level signal by weighted fusion of the reduced indexes.

[0006] Optionally, the multi-source data fusion module comprises a satellite remote sensing data acquisition unit, an ocean sensor network data acquisition unit, a meteorological station data acquisition unit, and a space-time alignment unit; wherein: The satellite remote sensing data acquisition unit is configured to receive satellite remote sensing data from a satellite remote sensing system in real time through a satellite sensor, and obtain ocean surface temperature, salinity, and sea surface height. The ocean sensor network data acquisition unit is configured to collect ocean sensor data including water temperature, dissolved oxygen, phytoplankton biomass, and chlorophyll concentration in real time through a sensor network arranged in the ocean. The meteorological station data acquisition unit is configured to obtain real-time meteorological station data including wind speed, air temperature, and humidity from a meteorological station. The space-time alignment unit is configured to perform space-time alignment processing on the satellite remote sensing data, the ocean sensor network data, and the meteorological station data, and unify the timestamps and geographical location information of different data sources to make the data consistent in time and space, thereby generating standardized ecological data streams.

[0007] Optionally, the dynamic feature extraction module comprises a data receiving unit, a sliding window mechanism unit, a change gradient calculation unit, and a result output unit; wherein: The data receiving unit is configured to receive standardized ecological data streams output from the multi-source data fusion module, including time series data of dissolved oxygen concentration, chlorophyll concentration, and phytoplankton biomass. The sliding window mechanism unit is configured to apply a sliding window mechanism to the received standardized ecological data streams, and set the window size to 72 hours, with the sliding window moving gradually in the time series. The change gradient calculation unit is configured to calculate the change rate of the three core parameters of dissolved oxygen concentration, chlorophyll concentration, and phytoplankton biomass within each sliding window, and obtain the change gradient of each parameter within 72 hours. The result output unit is configured to output the 72-hour change gradient of the three core parameters calculated in each sliding window.

[0008] Optionally, the change gradient calculation unit comprises: The dissolved oxygen concentration change rate calculation sub-unit is configured to receive dissolved oxygen concentration data from the sliding window mechanism unit, and obtain the change gradient of the dissolved oxygen concentration by solving the ratio of the change value of the dissolved oxygen concentration per hour to the time interval. The chlorophyll concentration change rate calculation sub-unit is configured to receive chlorophyll concentration data from the sliding window mechanism unit, and obtain the change gradient of the chlorophyll concentration by differentiating the chlorophyll concentration change at each time point and calculating the change rate according to the time interval. Plankton biomass change rate calculation subunit: for receiving the plankton biomass data from the sliding window mechanism unit, obtaining the change gradient of the plankton biomass by calculating the difference of the plankton biomass between time points and the ratio of the time interval; Gradient integration subunit: for integrating the change rates calculated by the three subunits to generate the change gradient of each parameter within 72 hours.

[0009] Optionally, the meteorological disturbance coefficient calculation module comprises a historical disaster data receiving unit, a meteorological disturbance sensitivity matrix construction unit and a meteorological disturbance coefficient calculation unit; wherein: The historical disaster data receiving unit is configured to receive meteorological disturbance data in a historical disaster database, including historical meteorological disaster events related to marine ecology; The meteorological disturbance sensitivity matrix construction unit is configured to establish a meteorological disturbance sensitivity matrix based on the meteorological disturbance data; The meteorological disturbance coefficient calculation unit is configured to calculate and output a meteorological disturbance coefficient according to the meteorological disturbance sensitivity matrix and real-time meteorological station data .

[0010] Optionally, the ecological index generation module comprises a water quality deterioration index generation unit and a biological stress index generation unit; wherein: The water quality deterioration index generation unit is configured to receive change gradient data of the dissolved oxygen concentration and the chlorophyll concentration, calculate the negative correlation between the dissolved oxygen concentration and the chlorophyll concentration, obtain the correlation strength, and generate a water quality deterioration index; The biological stress index generation unit is configured to receive change gradient data of the chlorophyll concentration and the plankton biomass, analyze the correlation between the chlorophyll concentration and the plankton biomass, obtain the positive correlation strength, and generate a biological stress index.

[0011] Optionally, the water quality deterioration index generation unit comprises: The data receiving subunit is configured to receive change gradient data of the dissolved oxygen concentration and the chlorophyll concentration; The correlation calculation subunit is configured to calculate the negative correlation coefficient between the dissolved oxygen concentration and the chlorophyll concentration; The correlation strength calculation subunit is configured to obtain the correlation strength between the dissolved oxygen concentration and the chlorophyll concentration based on the calculated negative correlation coefficient, and standardize it to a value between 0 and 1; The water quality deterioration index generation subunit is configured to obtain the final water quality deterioration index according to weighted calculation of the correlation strength, and the formula is: wherein, WQI is the water quality deterioration index; is a coefficient; represents the negative correlation strength between the dissolved oxygen concentration and the chlorophyll concentration.

[0012] Optionally, the biological stress index generating unit comprises: a data receiving subunit configured to receive data of variation gradient of chlorophyll concentration and phytoplankton biomass; a correlation calculating subunit configured to analyze the correlation between chlorophyll concentration and phytoplankton biomass, and quantify the relationship strength therebetween by calculating the Pearson correlation coefficient therebetween; a correlation strength calculating subunit configured to derive the positive correlation strength between chlorophyll concentration and phytoplankton biomass based on the calculation result of the Pearson correlation coefficient, and normalize the positive correlation strength to a value between 0 and 1; a biological stress index generating subunit configured to weight the biological stress index according to the size of the correlation strength, and finally derive the biological stress index, with the formula being: wherein BSI is the biological stress index, is a coefficient, and the positive correlation strength between chlorophyll concentration and phytoplankton biomass.

[0013] Optionally, the coupling early warning generating module comprises a meteorological interference component reduction unit, a weighted fusion unit and an ecological risk level signal generating unit; wherein: the meteorological interference component reduction unit is configured to, when the meteorological disturbance coefficient exceeds a preset threshold, calculate the meteorological interference component to be reduced according to the comparison result of the meteorological disturbance coefficient and the preset threshold, and then adjust the water quality deterioration index and the biological stress index, with the formula being: ; ; wherein, is an adjustment coefficient; is the maximum value of the meteorological disturbance coefficient; is the meteorological disturbance coefficient; is the preset threshold; the weighted fusion unit is configured to weight the reduced water quality deterioration index and the biological stress index, and calculate the ecological risk value ERS; the ecological risk level signal generating unit is configured to receive the ecological risk value ERS, and map it to the corresponding ecological risk level signal according to a preset rule.

[0014] Optionally, the mapping to the corresponding ecological risk level signal according to the preset rule comprises: when the ecological risk value ERS is less than or equal to 0.3, a low risk signal is generated; when the ecological risk value ERS is greater than 0.3 and less than or equal to 0.5, a low-medium risk signal is generated; When the ecological risk value ERS is greater than 0.5 and less than or equal to 0.7, a medium-risk signal is generated; When the ecological risk value ERS is greater than 0.7 and less than or equal to 0.85, a medium-to-high risk signal is generated. When the Ecological Risk Value (ERS) is greater than 0.85, a high-risk signal is generated.

[0015] The beneficial effects of this invention are: This invention, by reducing meteorological interference components and combining a weighted fusion of water quality deterioration index and biological stress index, can effectively identify and assess the impact of meteorological factors on the marine ecological environment. Through real-time adjustment of indicator weights, the system can accurately reflect the changing trends of the marine ecological environment under different meteorological conditions, providing a dynamic and comprehensive risk assessment, and solving the problem that existing technologies cannot process meteorological impacts and provide early warnings in real time. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Fig. 1 This is a schematic diagram of a marine ecological environment assessment and early warning system according to an embodiment of the present invention; Fig. 2 This is a schematic diagram of the coupling early warning generation module in an embodiment of the present invention. Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0019] like Figs. 1-2 As shown, a marine ecological environment assessment and early warning system based on big data and AI includes a multi-source data fusion module, a dynamic feature extraction module, a meteorological disturbance coefficient calculation module, an ecological indicator generation module, and a coupled early warning generation module; wherein: Multi-source data fusion module: used to collect satellite remote sensing data, marine sensor network data and meteorological station data in real time, and generate standardized ecological data streams through spatiotemporal alignment processing; Dynamic feature extraction module: used to receive standardized ecological data streams and use a sliding window mechanism to extract the 72-hour variation gradients of three core parameters: dissolved oxygen concentration, chlorophyll concentration, and plankton biomass; Meteorological disturbance coefficient calculation module: Based on the meteorological disturbance sensitivity matrix established by the historical disaster database, it calculates and outputs the meteorological disturbance coefficient; Ecological index generation module: It is used to receive the 72-hour change gradient of three core parameters, generate a water quality deterioration index based on the negative correlation between dissolved oxygen concentration and chlorophyll concentration, and generate a biological stress index based on the positive correlation between chlorophyll concentration and plankton biomass. Coupled early warning generation module: When the meteorological disturbance coefficient exceeds the preset threshold, it proportionally reduces the meteorological disturbance component in the water quality deterioration index and the biological stress index, and then performs weighted fusion on the reduced index to output the ecological risk level signal.

[0020] The multi-source data fusion module includes a satellite remote sensing data acquisition unit, a marine sensor network data acquisition unit, a meteorological station data acquisition unit, and a spatiotemporal alignment unit; among which: Satellite remote sensing data acquisition unit: used to receive satellite remote sensing data from the satellite remote sensing system in real time through satellite sensors, and to obtain ocean surface temperature, salinity and sea level height; Marine sensor network data acquisition unit: used to collect marine sensor data in real time, including parameters such as water temperature, dissolved oxygen, plankton biomass, and chlorophyll concentration, through a sensor network deployed in the ocean; Meteorological station data acquisition unit: used to acquire real-time meteorological station data from meteorological stations, including wind speed, temperature and humidity, to ensure comprehensive environmental data support; Spatiotemporal alignment unit: Used to perform spatiotemporal alignment processing on satellite remote sensing data, marine sensor network data, and meteorological station data. By uniformly processing the timestamps and geographic location information of different data sources, the data achieves consistency in time and space, thereby generating a standardized ecological data stream and ensuring that data from different sources can be effectively compared and analyzed. Through the combination of multi-source data acquisition units, the system can collect data from satellite remote sensing, marine sensor networks, and meteorological stations in real time, providing comprehensive marine ecological monitoring data. Spatiotemporal alignment ensures the consistency of data in time and space, thereby realizing the generation of a standardized ecological data stream and providing a reliable data foundation for subsequent feature extraction and evaluation.

[0021] The dynamic feature extraction module includes a data receiving unit, a sliding window mechanism unit, a gradient calculation unit, and a result output unit; wherein: Data receiving unit: for receiving standardized ecological data stream from multi-source data fusion module output, including time series data of dissolved oxygen concentration, chlorophyll concentration and plankton biomass; Sliding window mechanism unit: for applying sliding window mechanism to the received standardized ecological data stream, setting the window size to 72 hours, and moving the sliding window on the time series, and calculating the change gradient of the three core parameters through the data in each window; Change gradient calculation unit: for calculating the change rate of the three core parameters of dissolved oxygen concentration, chlorophyll concentration and plankton biomass in each sliding window, and obtaining the change gradient of each parameter within 72 hours; Result output unit: for outputting the 72-hour change gradient of the three core parameters calculated by each sliding window, for use by the subsequent evaluation module; through the sliding window mechanism, the dynamic feature extraction module can efficiently extract the change gradient of the key ecological parameters within 72 hours from the standardized ecological data stream; this method can capture the dynamic change trend of the marine ecological environment in real time, providing high-precision input data for subsequent ecological risk assessment and early warning, and helping to identify environmental changes and make scientific predictions in a timely manner.

[0022] The change gradient calculation unit includes: Dissolved oxygen concentration change rate calculation sub-unit: for receiving dissolved oxygen concentration data from the sliding window mechanism unit, obtaining the change gradient of dissolved oxygen concentration by solving the ratio of the change value of dissolved oxygen concentration per hour to the time interval; Chlorophyll concentration change rate calculation sub-unit: for receiving chlorophyll concentration data from the sliding window mechanism unit, obtaining the change gradient of chlorophyll concentration by differentiating the chlorophyll concentration change at each time point and calculating the change rate according to the time interval; Plankton biomass change rate calculation sub-unit: for receiving plankton biomass data from the sliding window mechanism unit, obtaining the change gradient of plankton biomass by calculating the ratio of the difference of plankton biomass at each time point to the time interval; Gradient integration sub-unit: for integrating the change rates calculated by the three sub-units to generate the change gradient of each parameter within 72 hours for use by the subsequent module; by calculating and integrating the change rates of the three core parameters one by one, the change gradient calculation unit can accurately extract the change trend of dissolved oxygen concentration, chlorophyll concentration and plankton biomass, and this function ensures that the dynamic changes of each parameter can be quantified scientifically and accurately, providing detailed data support for subsequent ecological assessment.

[0023] The meteorological disturbance coefficient calculation module includes a historical disaster data receiving unit, a meteorological disturbance sensitivity matrix construction unit, and a meteorological disturbance coefficient calculation unit; wherein: Historical disaster data receiving unit: for receiving meteorological disturbance data in the historical disaster database, including historical meteorological disaster events related to marine ecology (such as storms, typhoons, etc.); Meteorological disturbance sensitivity matrix construction unit: based on meteorological disturbance data, establish meteorological disturbance sensitivity matrix; The steps of constructing the meteorological disturbance sensitivity matrix are as follows: Data collection: collect historical disaster data, including typhoons, storms, droughts, heat waves, etc.; Collect relevant ecological environment data, including data of target ecological parameters such as dissolved oxygen concentration, chlorophyll concentration, and plankton biomass; Select meteorological factors: determine the main meteorological factors that affect the ecological environment, including wind speed, air pressure, precipitation, temperature, etc.; Determine the influence: first, analyze the influence of each meteorological factor on the target ecological environment parameter according to the historical disaster data and ecological environment data; then use statistical analysis methods to quantify the relationship strength between meteorological factors and target parameters; finally, generate an influence value to represent the influence strength of meteorological factors on ecological environment parameters, which can be represented by a value between 0 and 1; Construct sensitivity matrix: organize the relationship between all meteorological factors and target ecological environment parameters into a matrix, each element of the matrix represents the influence sensitivity of a meteorological factor on a target ecological parameter; the rows of the matrix represent different meteorological factors, and the columns represent different ecological parameters; Standardization: standardize each influence degree to ensure that the elements in the matrix have a relatively uniform scale; Final matrix: finally obtain the meteorological disturbance sensitivity matrix, which can reflect the influence strength of different meteorological factors on the ecological environment.

[0024] There are n meteorological factors and m ecological environment parameters, and the meteorological disturbance sensitivity matrix S is an n × m matrix, each element of the matrix represents the influence strength of the i-th meteorological factor on the j-th ecological environment parameter; its expression is as follows: ; Wherein, each value is obtained by actual data analysis; the sensitivity value can be positive (indicating positive correlation) or negative (indicating negative correlation), and the higher the sensitivity, the greater the influence of the change of the meteorological factor on the ecological environment parameter.

[0025] Meteorological disturbance coefficient calculation unit: for calculating the output meteorological disturbance coefficient according to the meteorological disturbance sensitivity matrix and the real-time collected meteorological station data; the calculation formula is as follows: , wherein, is the sensitivity of the weather parameter i to the ecosystem parameter j, and is the real-time change value of the weather parameter i (such as current wind speed, precipitation, etc.); is the weather disturbance coefficient; n is the total number of weather parameters.

[0026] The ecological index generation module includes a water quality deterioration index generation unit and a biological stress index generation unit; wherein: The water quality deterioration index generation unit is configured to receive variation gradient data of dissolved oxygen concentration and chlorophyll concentration, calculate the negative correlation between the dissolved oxygen concentration and the chlorophyll concentration, obtain the correlation strength between them, and generate a water quality deterioration index; The biological stress index generation unit is configured to receive variation gradient data of chlorophyll concentration and plankton biomass, analyze the correlation between the chlorophyll concentration and the plankton biomass, obtain the positive correlation strength between them, and generate a biological stress index; by calculating the water quality deterioration index and the biological stress index, the ecological index generation module can assess the trend of water quality deterioration and ecological stress in real time, provide effective data support, help to identify changes in the marine ecological environment in a timely manner, and provide a scientific basis for subsequent risk warning and management decision-making.

[0027] The water quality deterioration index generation unit includes: A data receiving subunit configured to receive variation gradient data of dissolved oxygen concentration and chlorophyll concentration; A correlation calculation subunit configured to calculate the negative correlation coefficient between the dissolved oxygen concentration and the chlorophyll concentration, quantify the relationship between them by calculating the Pearson correlation coefficient, and the formula is: wherein, and represent the values of the dissolved oxygen concentration and the chlorophyll concentration at the i-th time point, and are the mean values of the dissolved oxygen concentration and the chlorophyll concentration, and r is the negative correlation coefficient between them; the negative correlation is indicated by the sign of the result value, and a negative value indicates a negative correlation between the dissolved oxygen concentration and the chlorophyll concentration; A correlation strength calculation subunit obtains the correlation strength between the dissolved oxygen concentration and the chlorophyll concentration based on the calculated negative correlation coefficient, normalizes it to a value between 0 and 1, and a higher strength indicates a more obvious negative correlation; A water quality deterioration index generation subunit obtains the final water quality deterioration index by weighted calculation according to the size of the correlation strength, and the formula is: wherein, WQI is the water quality deterioration index; is a coefficient for adjusting the scale of the index value; represents the negative correlation intensity between the dissolved oxygen concentration and the chlorophyll concentration.

[0028] The biological stress index generation unit comprises: a data receiving subunit for receiving data of the variation gradient of the chlorophyll concentration and the phytoplankton biomass; a correlation calculation subunit for analyzing the correlation between the chlorophyll concentration and the phytoplankton biomass, quantifying the relationship intensity therebetween by calculating the Pearson correlation coefficient therebetween; a correlation intensity calculation subunit for deriving the positive correlation intensity between the chlorophyll concentration and the phytoplankton biomass based on the calculation result of the Pearson correlation coefficient, normalizing the positive correlation intensity to a value between 0 and 1, and the higher the intensity, the more obvious the positive correlation between the two; a biological stress index generation subunit for calculating the biological stress index according to the size of the correlation intensity, and finally deriving the biological stress index, the formula being: wherein BSI is the biological stress index, is a coefficient for adjusting the scale of the index value, represents the positive correlation intensity between the chlorophyll concentration and the phytoplankton biomass; by calculating the positive correlation between the chlorophyll concentration and the phytoplankton biomass and generating the biological stress index based thereon, the biological stress index generation unit can accurately evaluate the mutual relationship between the algal growth and the phytoplankton in the ecosystem, and provide a scientific basis for the monitoring and early warning of biological stress.

[0029] The coupling early warning generation module comprises a meteorological interference component reduction unit, a weighted fusion unit and an ecological risk level signal generation unit; wherein: The meteorological interference component reduction unit is configured to, when the meteorological disturbance coefficient exceeds the preset threshold, calculate the meteorological interference component to be reduced according to the comparison result of the meteorological disturbance coefficient and the preset threshold, and then adjust the water quality deterioration index and the biological stress index; the calculation formula is: ; ; wherein, is an adjustment coefficient; is the maximum value of the meteorological disturbance coefficient; is the meteorological disturbance coefficient; is the preset threshold; this process reduces the meteorological influence component in the water quality deterioration index and the biological stress index by the proportional reduction method; The weighted fusion unit is configured to perform weighted fusion on the reduced water quality deterioration index and the biological stress index to calculate the ecological risk value ERS; the weighted fusion is performed according to the set weight coefficients and to calculate the ecological risk value by weighting the two adjusted indexes, and the formula is as follows: wherein, and are the weight coefficients of water quality deterioration index and biological stress index respectively, satisfying to ensure that the result of weighted fusion reasonably reflects the relative importance of the two indicators; According to the current dissolved oxygen concentration, the weight is dynamically allocated: When the dissolved oxygen concentration is ≥5 mg / L, the water quality deterioration weight =0.6, and the biological stress weight =0.4; When the dissolved oxygen concentration is <5 mg / L, the water quality deterioration weight =0.3, and the biological stress weight =0.7; The ecological risk level signal generation unit is used to receive the ecological risk value ERS and map it to the corresponding ecological risk level signal according to the preset rules; the above unit can accurately assess the risk level of the marine ecological environment by eliminating meteorological interference components and weightedly fusing the water quality deterioration index and the biological stress index, coupled with the early warning generation module; according to the real-time data, the weight is dynamically adjusted, and the generated ecological risk level signal provides scientific support for marine ecological protection and pollution early warning, which helps to realize early warning and effective response.

[0030] Mapping it to the corresponding ecological risk level signal according to the preset rules includes: When the ecological risk value ERS is less than or equal to 0.3, it indicates that the ecological environment is in a low-risk state, and a low-risk signal is generated; When the ecological risk value ERS is greater than 0.3 and less than or equal to 0.5, it indicates that the ecological environment is in a low-to-medium risk state, and a low-to-medium risk signal is generated; When the ecological risk value ERS is greater than 0.5 and less than or equal to 0.7, it indicates that the ecological environment is in a medium risk state, and a medium risk signal is generated; When the ecological risk value ERS is greater than 0.7 and less than or equal to 0.85, it indicates that the ecological environment is in a high-to-medium risk state, and a high-to-medium risk signal is generated; When the ecological risk value ERS is greater than 0.85, it indicates that the ecological environment is in a high-risk state, and a high-risk signal is generated; by mapping the ecological risk value to a specific risk level signal, the ecological risk level signal generation unit can provide clear risk indication to help decision-makers understand the risk level of the marine ecological environment in real time; these risk level signals provide a strong basis for ecological environment management and early warning.

[0031] The present application encompasses any alternatives, modifications, equivalent methods and solutions made to the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can also be fully understood without the description of these details to those skilled in the art. In addition, in order to avoid unnecessary confusion to the essence of the present application, well-known methods, processes, procedures, elements and circuits, etc. are not described in detail.

[0032] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can also be made, which should be considered as the protection scope of the present application.

Claims

1. A marine ecological environment assessment and early warning system based on big data and AI, characterized in that, It includes a multi-source data fusion module, a dynamic feature extraction module, a meteorological disturbance coefficient calculation module, an ecological indicator generation module, and a coupled early warning generation module; among which: Multi-source data fusion module: used to collect satellite remote sensing data, marine sensor network data and meteorological station data in real time, and generate standardized ecological data streams through spatiotemporal alignment processing; Dynamic feature extraction module: used to receive standardized ecological data streams and use a sliding window mechanism to extract the 72-hour variation gradients of three core parameters: dissolved oxygen concentration, chlorophyll concentration, and plankton biomass; Meteorological disturbance coefficient calculation module: Based on the meteorological disturbance sensitivity matrix established by the historical disaster database, it calculates and outputs the meteorological disturbance coefficient; Ecological index generation module: It is used to receive the 72-hour change gradient of three core parameters, generate a water quality deterioration index based on the negative correlation between dissolved oxygen concentration and chlorophyll concentration, and generate a biological stress index based on the positive correlation between chlorophyll concentration and plankton biomass. Coupled early warning generation module: When the meteorological disturbance coefficient exceeds the preset threshold, it proportionally reduces the meteorological disturbance component in the water quality deterioration index and the biological stress index, and then performs weighted fusion on the reduced index to output the ecological risk level signal.

2. The marine ecological environment assessment and early warning system based on big data and AI according to claim 1, characterized in that, The multi-source data fusion module includes a satellite remote sensing data acquisition unit, a marine sensor network data acquisition unit, a meteorological station data acquisition unit, and a spatiotemporal alignment unit; wherein: Satellite remote sensing data acquisition unit: used to receive satellite remote sensing data from the satellite remote sensing system in real time through satellite sensors, and to obtain ocean surface temperature, salinity and sea level height; Marine sensor network data acquisition unit: used to collect marine sensor data in real time, including parameters such as water temperature, dissolved oxygen, plankton biomass, and chlorophyll concentration, through a sensor network deployed in the ocean; Meteorological station data acquisition unit: used to acquire real-time meteorological station data from meteorological stations, including wind speed, temperature and humidity; Spatiotemporal alignment unit: used to perform spatiotemporal alignment processing on satellite remote sensing data, marine sensor network data and meteorological station data. By uniformly processing the timestamps and geographic location information of different data sources, the data is made consistent in time and space, thereby generating a standardized ecological data stream.

3. The marine ecological environment assessment and early warning system based on big data and AI according to claim 1, characterized in that, The dynamic feature extraction module includes a data receiving unit, a sliding window mechanism unit, a gradient calculation unit, and a result output unit; wherein: Data receiving unit: used to receive standardized ecological data streams output from the multi-source data fusion module, including time-series data of dissolved oxygen concentration, chlorophyll concentration and plankton biomass; Sliding window mechanism unit: used to apply the sliding window mechanism to the received standardized ecological data stream, with the window size set to 72 hours, and the sliding window moving gradually over time. The gradient calculation unit is used to calculate the rate of change of three core parameters—dissolved oxygen concentration, chlorophyll concentration, and plankton biomass—within each sliding window, and obtain the gradient of change of each parameter over 72 hours. Results output unit: Used to output the 72-hour gradient of the three core parameters calculated for each sliding window.

4. The marine ecological environment assessment and early warning system based on big data and AI according to claim 3, characterized in that, The gradient calculation unit includes: Dissolved oxygen concentration change rate calculation subunit: It is used to receive dissolved oxygen concentration data from the sliding window mechanism unit, and obtain the gradient of dissolved oxygen concentration change by solving the ratio of the hourly dissolved oxygen concentration change value to the time interval. Chlorophyll concentration change rate calculation subunit: It is used to receive chlorophyll concentration data from the sliding window mechanism unit, and obtain the chlorophyll concentration change gradient by differentiating the chlorophyll concentration change at each time point and calculating the change rate according to the time interval. The phytoplankton biomass change rate calculation subunit is used to receive phytoplankton biomass data from the sliding window mechanism unit and obtain the phytoplankton biomass change gradient by calculating the ratio of the difference in phytoplankton biomass between each time point to the time interval. Gradient Integration Subunit: Used to integrate the rate of change calculated from the three subunits to generate the gradient of change of each parameter over 72 hours.

5. A marine ecological environment assessment and early warning system based on big data and AI according to claim 1, characterized in that, The meteorological disturbance coefficient calculation module includes a historical disaster data receiving unit, a meteorological disturbance sensitivity matrix construction unit, and a meteorological disturbance coefficient calculation unit; wherein: Historical disaster data receiving unit: used to receive meteorological disturbance data from the historical disaster database, including historical meteorological disaster events related to marine ecology; Meteorological disturbance sensitivity matrix construction unit: Based on meteorological disturbance data, a meteorological disturbance sensitivity matrix is ​​established; Meteorological Disturbance Coefficient Calculation Unit: Used to calculate and output the meteorological disturbance coefficient based on the meteorological disturbance sensitivity matrix and real-time collected meteorological station data. .

6. The marine ecological environment assessment and early warning system based on big data and AI according to claim 1, characterized in that, The ecological indicator generation module includes a water quality deterioration index generation unit and a biological stress index generation unit; wherein: Water quality deterioration index generation unit: It is used to receive gradient data of changes in dissolved oxygen concentration and chlorophyll concentration, calculate the negative correlation between dissolved oxygen concentration and chlorophyll concentration, determine their correlation strength, and thus generate a water quality deterioration index. Biological stress index generation unit: It is used to receive gradient data of changes in chlorophyll concentration and plankton biomass, analyze the correlation between chlorophyll concentration and plankton biomass, determine the strength of their positive correlation, and thus generate a biological stress index.

7. A marine ecological environment assessment and early warning system based on big data and AI according to claim 6, characterized in that, The water quality deterioration index generation unit includes: Data receiving subunit: used to receive gradient data of changes in dissolved oxygen concentration and chlorophyll concentration; Correlation calculation subunit: used to calculate the negative correlation coefficient between dissolved oxygen concentration and chlorophyll concentration; Correlation strength calculation subunit: Based on the calculated negative correlation coefficient, the correlation strength between dissolved oxygen concentration and chlorophyll concentration is obtained and standardized to a value between 0 and 1; Water quality deterioration index generation sub-unit: The final water quality deterioration index is calculated by weighting the values ​​based on the correlation strength, using the following formula: Among them, WQI is the water quality deterioration index; For coefficients; This indicates the strength of the negative correlation between dissolved oxygen concentration and chlorophyll concentration.

8. A marine ecological environment assessment and early warning system based on big data and AI according to claim 6, characterized in that, The biological stress index generation unit includes: Data receiving subunit: used to receive gradient data on changes in chlorophyll concentration and plankton biomass; Correlation calculation subunit: used to analyze the correlation between chlorophyll concentration and plankton biomass, and quantifies the strength of the relationship between the two by calculating the Pearson correlation coefficient; Correlation strength calculation subunit: Based on the calculation results of Pearson correlation coefficient, the positive correlation strength between chlorophyll concentration and plankton biomass is obtained and standardized to a value between 0 and 1; Biological stress index generation sub-unit: The biological stress index is calculated by weighting the values ​​of the correlation strengths to obtain the final biological stress index, as shown in the formula: Among them, BSI is the biological stress index. For coefficients, This indicates the strength of the positive correlation between chlorophyll concentration and plankton biomass.

9. A marine ecological environment assessment and early warning system based on big data and AI according to claim 1, characterized in that, The coupled early warning generation module includes a meteorological interference component reduction unit, a weighted fusion unit, and an ecological risk level signal generation unit; wherein: Meteorological disturbance component reduction unit: When the meteorological disturbance coefficient exceeds a preset threshold, it calculates the meteorological disturbance component that needs to be reduced based on the comparison between the meteorological disturbance coefficient and the preset threshold, and then adjusts the water quality deterioration index and biological stress index; the calculation formula is: ； ;in, It is an adjustment factor; This represents the maximum value of the meteorological disturbance coefficient; This is the meteorological disturbance coefficient; This is a preset threshold value; Weighted fusion unit: used to weight and fuse the reduced water quality deterioration index and biological stress index to calculate the ecological risk value ERS; Ecological risk level signal generation unit: used to receive ecological risk value ERS and map it to the corresponding ecological risk level signal according to preset rules.

10. A marine ecological environment assessment and early warning system based on big data and AI according to claim 9, characterized in that, The process of mapping the data to the corresponding ecological risk level signal according to preset rules includes: When the ecological risk value ERS is less than or equal to 0.3, a low-risk signal is generated; When the ecological risk value ERS is greater than 0.3 and less than or equal to 0.5, a medium-to-low risk signal is generated. When the ecological risk value ERS is greater than 0.5 and less than or equal to 0.7, a medium-risk signal is generated; When the ecological risk value ERS is greater than 0.7 and less than or equal to 0.85, a medium-to-high risk signal is generated. When the Ecological Risk Value (ERS) is greater than 0.85, a high-risk signal is generated.

Images