CN-121981392-A - Marine ecology evaluation analysis system based on big data

CN121981392ACN 121981392 ACN121981392 ACN 121981392ACN-121981392-A

Abstract

The invention discloses a marine ecological evaluation analysis system based on big data, which relates to the technical field of marine environment monitoring and information and comprises a data sensing acquisition module, a characteristic analysis extraction module, an ecological parameter mapping module, a self-adaptive multi-level evaluation module and an ecological diagnosis report module. The system acquires the original data of the marine environment through a sensing network, and converts the original data into parameters to be evaluated with ecological meaning through processing. The self-adaptive evaluation module can dynamically adjust the execution path and the strength of the evaluation strategy according to the data characteristics and the environmental context of the input parameters, and then performs composite calculation and fusion analysis. And the reporting module longitudinally compares the current quantitative diagnosis conclusion with the historical evaluation record, identifies the ecological state change track and generates a structured report. According to the scheme, the dynamic intelligent adaptation of the evaluation strategy and the deep mining of the ecological evolution trend are realized, so that the accuracy of an evaluation result is effectively improved, and the decision support capability of long-term ecological monitoring and management is enhanced.

Inventors

Qing Haihang
ZHANG WEIFENG
LI XIN
LIU YUAN
XU LI
LIU HE
YANG MENGCHEN
ZHOU CHAO

Assignees

国科检测技术服务(山东)有限公司

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (10)

1. A marine ecology evaluation analysis system based on big data, comprising: the data perception acquisition module is used for capturing original multi-source heterogeneous marine environment information by a perception network; the feature analysis and extraction module is used for analyzing and structuring the captured original marine environment information and extracting basic feature quantity for representing ecological conditions; The ecological parameter mapping module is used for mapping the basic characteristic quantity into a plurality of parameters to be evaluated with clear ecological meanings according to preset ecological rules and expert knowledge; The self-adaptive multi-level evaluation module is used for constructing a multi-level evaluation framework comprising a self-adaptive adjustment mechanism, inputting the multiple parameters to be evaluated into the multi-level evaluation framework, dynamically adjusting the execution path and the intensity of an evaluation strategy by the self-adaptive adjustment mechanism according to the data characteristics of the input parameters and the environmental context, and executing the adjusted evaluation strategy to perform composite calculation and fusion analysis on the parameters to be evaluated; And the ecological diagnosis report module is used for generating a quantitative diagnosis conclusion based on the fusion analysis result, longitudinally comparing the quantitative diagnosis conclusion with the historical evaluation record to identify the change track of the ecological state, and integrating the current diagnosis conclusion and the historical change track to form a structured ecological evaluation report.
2. The marine ecology evaluation analysis system based on big data of claim 1, wherein in the step of capturing original marine environment information of multi-source heterogeneous by the perception network, specifically comprising: Deploying a composite sensing array comprising physical sensor nodes and a remote sensing image receiving unit; the composite sensing array continuously captures multidimensional dynamic data streams including a temperature field, a chemical substance concentration field, bioacoustic signals and spectral reflection characteristics from a seawater medium, an atmospheric interface and a seabed substrate in an asynchronous parallel mode; performing timestamp synchronization and space coordinate registration on the multidimensional dynamic data stream, and eliminating data dislocation caused by asynchronous acquisition source and acquisition time; and packaging the multidimensional dynamic data stream with synchronization and registration completion into an original information data packet with a unified space-time reference, wherein the original information data packet is the original marine environment information.
3. The marine ecology evaluation analysis system based on big data of claim 2, wherein in the step of analyzing and structuring the captured original marine environment information, the specific steps are: Receiving an original information data packet from the perception network; noise filtering and abnormal value detection are carried out on the multidimensional dynamic data flow in the original information data packet, and invalid data fragments generated by equipment failure or instantaneous interference are removed; Identifying and separating a change mode, periodic fluctuation and trend components which can reflect a key ecological process from the cleaned data stream by adopting a characteristic extraction algorithm; And respectively quantifying the change mode, the periodic fluctuation and the trend component into numerical value type sequences, and marking the corresponding ecological process types, thereby forming a group of basic characteristic quantities with standard formats and process directivity.
4. A marine data based evaluation analysis system according to claim 3, characterized in that in the link of mapping the basic feature quantity into a plurality of parameters to be evaluated with a clear ecological meaning, the following procedure is implemented: presetting an ecological semantic mapping knowledge base, wherein the ecological semantic mapping knowledge base stores association rules and weight coefficients between various basic feature quantities and ecological concepts; inputting the basic feature quantity of the group of standard formats into the ecological semantic mapping knowledge base; matching ecological concepts corresponding to each type of basic feature quantity by an association rule engine in the ecological semantic mapping knowledge base, and carrying out weighted aggregation on contributions of multiple types of feature quantities according to weight coefficients; And outputting an aggregation result, wherein the aggregation result is presented in the form of a parameter set, namely a parameter set to be evaluated, each parameter represents an ecological state index which is comprehensively judged, and the ecological state index forms the parameter to be evaluated.
5. The marine ecology evaluation analysis system based on big data of claim 4 wherein the construction and input process of the multi-level evaluation framework comprising an adaptive adjustment mechanism comprises: Designing a frame body with a tree topology structure, wherein different levels of the frame body correspond to ecological evaluation dimensions with different granularities; Embedding the self-adaptive adjusting mechanism into a key decision node of the framework body, wherein the self-adaptive adjusting mechanism is internally provided with a plurality of evaluation strategy templates and selection logic; And importing the parameter set to be evaluated into a root node of the framework body, and distributing the parameters to downstream nodes along the tree topology structure.
6. The marine ecology evaluation analysis system based on big data of claim 5 wherein the process of dynamically adjusting the execution path and intensity of the evaluation strategy by the adaptive adjustment mechanism according to the data characteristics of the input parameters and the environmental context is: the self-adaptive adjustment mechanism receives parameter distribution state data and environment context metadata from each node of the framework body in real time; analyzing the parameter distribution state data, and judging the completeness and confidence of the data under each evaluation dimension; Meanwhile, analyzing the environmental context metadata, and identifying the specific scene type and constraint condition of the current evaluation task; and dynamically selecting the most adapted strategy combination from a plurality of evaluation strategy templates according to a preset adaptation rule by taking data completeness, confidence coefficient, scene type and constraint condition as inputs, and calculating the execution intensity coefficient of the strategy combination in the current context, thereby completing the adjustment of the execution path and intensity of the evaluation strategy.
7. The marine ecology evaluation analysis system based on big data of claim 6 wherein the process of performing composite calculation and fusion analysis on the parameter to be evaluated is: executing strategy combination selected by the self-adaptive adjustment mechanism in the multi-level evaluation framework; performing cross-dimension correlation analysis, time sequence comparison and space superposition operation on the parameters to be evaluated of different levels according to the algorithm flow defined by the strategy combination; performing evidence synthesis and uncertainty reduction on intermediate results generated by association analysis, time sequence comparison and space superposition operation at a fusion center node of the framework; And finally outputting a comprehensive quantization score vector reflecting the overall ecological condition and an affiliated confidence interval thereof, wherein the quantization score vector and the confidence interval thereof are the fusion analysis result.
8. The big data based marine ecology evaluation analysis system of claim 7, wherein the step of generating a quantitative diagnosis conclusion based on a result of the fusion analysis is: analyzing a quantization score vector in the fusion analysis result; Comparing each score in the quantization score vector with a preset, graded health status threshold interval; According to the comparison result, a state grade label is given to each estimated ecological dimension or overall condition; summarizing all the state grade labels, and attaching the corresponding confidence intervals to form a quantitative diagnosis conclusion containing grading judgment and reliability measurement.
9. The big data based marine ecology evaluation analysis system of claim 8, wherein the process of longitudinally comparing the quantitative diagnostic conclusion with a historical evaluation record comprises: retrieving historical quantitative diagnosis conclusions of the target sea area at a plurality of past time points from a system storage; arranging the current generated quantitative diagnosis conclusion and the historical quantitative diagnosis conclusion in time sequence, and extracting a state grade label and a change sequence of a grading value of the current quantitative diagnosis conclusion and the historical quantitative diagnosis conclusion aiming at the same ecological dimension or index; Analyzing the change sequence, and identifying a trend segment with improved, degenerated or maintained stability of the index and a key time point at which state turning possibly occurs, so as to outline the change track of the ecological state.
10. The big data based marine ecology evaluation analysis system of claim 9, wherein the integrating the current diagnostic conclusion with the historical variation trace forms a final link of a structured ecology evaluation report comprising: creating a standardized report template containing abstracts, subject analyses and appendices; filling the current quantitative diagnosis conclusion generated at present into a current situation analysis part of a report body; filling the change track of the ecological state into a trend analysis part of a report body in a form of combining a chart with a text description; Recording a strategy selection log of a basic feature quantity, key parameters and an adaptive adjustment mechanism adopted by the evaluation in a report appendix; integrating all parts to generate a complete and traceable structured ecological evaluation report.

Description

Marine ecology evaluation analysis system based on big data Technical Field The invention belongs to the technical field of marine environment monitoring and information, and particularly relates to a marine ecological evaluation analysis system based on big data. Background Existing marine ecological assessment techniques typically rely on preset static assessment models. The models are based on a fixed index system, an algorithm flow and a weight threshold value, and the collected environmental data are calculated and analyzed. Once its evaluation strategy is set, it remains unchanged throughout the application process. When the quality, distribution or structure of the input data fluctuates or the specific environment background where the evaluation task is located changes, the static model cannot correspondingly adjust the internal processing logic. This stiff evaluation mode results in challenges in reliability and applicability of the output results of the system in the face of data anomalies or context changes, and there are significant limitations in the adaptability of the evaluation process. The current general evaluation system mainly focuses on the processing of the current acquired data and outputs the ecological condition conclusion or grade on the time section. The system generally lacks functional modules for automated, standardized associative alignment of current analysis results with historical archived data. Often, the manager obtains an isolated evaluation report, and it is difficult to directly and systematically identify the overall track, long-term trend and abnormal turning points of each ecological parameter evolving along with time from a series of reports. The understanding of the dynamic process of the ecological system is fragmented, and the supporting value of the evaluation information on long-term management and trend early warning is weakened. There is a need to develop a marine ecological assessment analysis system that can break through the limitations described above. One of the core problems to be solved by the system is to realize dynamic intelligent adaptation of an evaluation strategy, so that the evaluation process can be automatically regulated according to real-time data characteristics and external environment context. The other core problem to be solved is to realize continuous trend analysis of ecological state, organically integrate the current diagnosis conclusion into the history sequence, and generate a structured report capable of revealing the evolution rule. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art; therefore, the invention provides a marine ecological evaluation analysis system based on big data, which comprises the following steps: the data perception acquisition module is used for capturing original multi-source heterogeneous marine environment information by a perception network; the feature analysis and extraction module is used for analyzing and structuring the captured original marine environment information and extracting basic feature quantity for representing ecological conditions; The ecological parameter mapping module is used for mapping the basic characteristic quantity into a plurality of parameters to be evaluated with clear ecological meanings according to preset ecological rules and expert knowledge; The self-adaptive multi-level evaluation module is used for constructing a multi-level evaluation framework comprising a self-adaptive adjustment mechanism, inputting the multiple parameters to be evaluated into the multi-level evaluation framework, dynamically adjusting the execution path and the intensity of an evaluation strategy by the self-adaptive adjustment mechanism according to the data characteristics of the input parameters and the environmental context, and executing the adjusted evaluation strategy to perform composite calculation and fusion analysis on the parameters to be evaluated; And the ecological diagnosis report module is used for generating a quantitative diagnosis conclusion based on the fusion analysis result, longitudinally comparing the quantitative diagnosis conclusion with the historical evaluation record to identify the change track of the ecological state, and integrating the current diagnosis conclusion and the historical change track to form a structured ecological evaluation report. Preferably, in the step of capturing the original multi-source heterogeneous marine environment information by the sensing network, the method specifically includes: Deploying a composite sensing array comprising physical sensor nodes and a remote sensing image receiving unit; the composite sensing array continuously captures multidimensional dynamic data streams including a temperature field, a chemical substance concentration field, bioacoustic signals and spectral reflection characteristics from a seawater medium, an atmospheric interface and a