CN-121810057-B - Marine ecological environment analysis method and system based on big data

CN121810057BCN 121810057 BCN121810057 BCN 121810057BCN-121810057-B

Abstract

The invention relates to the technical field of marine ecological environment analysis and discloses a marine ecological environment analysis method and a marine ecological environment analysis system based on big data, wherein the method comprises the steps of acquiring environment monitoring and control object working condition time sequence data, and determining a transmission lag time window based on flow field transmission logic; the system comprises a data acquisition and space-time mapping module, a cross-correlation calculation module, an active attribution right determination module and a fingerprint verification and decision module, wherein the system realizes exclusive locking of a responsible main body in a complex hydrologic environment through an active pseudo-random disturbance and frequency domain orthogonal verification mechanism, and solves the attribution problem that passive monitoring is difficult to distinguish natural background fluctuation from artificial emission.

Inventors

ZHANG CHAO
LI ZHILIN
TAO HUIMIN
Qiu Shaonan
ZHANG XIAOWEN
LI HONGKUN
CHEN WEI
SU BO
ZHAO YUTING

Assignees

山东省海洋资源与环境研究院(山东省海洋环境监测中心、山东省水产品质量检验中心)

Dates

Publication Date: 20260512
Application Date: 20260310

Claims (7)

1. The marine ecological environment analysis method based on big data is characterized by comprising the following steps of: acquiring environment monitoring time sequence data of a target sea area and working condition operation time sequence data of a control object, and determining a transmission lag time window from the control object to an environment monitoring point position based on a preset flow field transmission logic; The method comprises the steps of calculating cross-correlation coefficients of working condition operation time sequence data and environment monitoring time sequence data in a transmission lag time window, responding to the cross-correlation coefficients to fall into a preset gray scale verification interval, generating a pseudo-random working condition control sequence, constructing frequency domain characteristics of the pseudo-random working condition control sequence to be orthogonal with background main frequency of current environment monitoring time sequence data, dynamically configuring the lower limit value of the gray scale verification interval based on a risk weighted sensitivity drift mechanism, specifically comprising the steps of establishing a risk accumulation account of a control object, extracting a historical cross-correlation coefficient of the control object when a supervision punishment instruction is not triggered in a preset historical period, screening values falling into a preset observation interval and calculating corresponding risk accumulation indexes, calculating the risk accumulation indexes into the risk accumulation account, and dynamically adjusting the lower limit value of the gray scale verification interval based on the current balance of the risk accumulation account, wherein the higher the current balance of the risk accumulation account is, the lower limit value of the gray scale verification interval is lower; The method comprises the steps of receiving a scheduling instruction containing a pseudo-random working condition control sequence from an administrative management interface, issuing the scheduling instruction containing the pseudo-random working condition control sequence to a management object through the administrative management interface, executing irregular action according to the pseudo-random working condition control sequence to enable the working condition running state of the management object to change correspondingly, generating the pseudo-random working condition control sequence, performing frequency spectrum analysis on environment monitoring time sequence data, extracting main environment background periodic frequency, constructing a notch filter logic on a frequency domain to inhibit a frequency band corresponding to the environment background periodic frequency, generating a broadband pseudo-random noise sequence, and filtering the broadband pseudo-random noise sequence by applying the notch filter logic to obtain the pseudo-random working condition control sequence; Acquiring verification environment time sequence data after irregular action execution, and calculating the local waveform matching degree of the verification environment time sequence data and the pseudo-random working condition control sequence; When the matching degree of local waveforms exceeds a right-confirming threshold value, judging that abnormal fluctuation of environment monitoring time sequence data is originated from a control object, establishing responsibility association mapping for the control object and generating a supervision punishment instruction, wherein a pseudo-random working condition control sequence is used for demodulating exclusive responsibility main body fingerprints from environment background noise under a multi-source interference environment, the step of establishing the responsibility association mapping adopts iterative residual error stripping logic, and concretely comprises the steps of determining a first-level responsibility association mapping based on the environment monitoring time sequence data, determining a first-level control object with the highest cross-correlation coefficient, generating a theoretical environment contribution sequence of the first-level control object based on the working condition operation time sequence data of the first-level control object and the corresponding cross-correlation coefficient, calculating the difference value between the environment monitoring time sequence data and the theoretical environment contribution sequence, and generating an environment abnormal residual error sequence The calculation logic thereof satisfies: , wherein, For the environmental monitoring of the time-series data, Operating time sequence data for the working condition of the first control object, For the response coefficients determined based on the cross-correlation coefficients, Constant of baseline, sequence of environment anomaly residuals As a new input reference, the calculation step of the cross-correlation coefficient is re-performed on the remaining management objects except the first management object to lock the secondary responsibility main body.
2. The marine ecological environment analysis method based on big data according to claim 1, further comprising the step of executing a data logic consistency check based on a side channel before the supervision punishment instruction is generated, wherein the step of acquiring energy consumption time sequence data of the control object during the irregular action, the energy consumption time sequence data are derived from a third party utility metering system, the step of calculating a logic synchronization rate between the energy consumption time sequence data and a pseudo-random working condition control sequence based on a preset pollution production and discharge coupling model, the step of blocking the generation of the supervision punishment instruction and generating a data audit work order for the control object in response to the logic synchronization rate being lower than a preset self-consistency threshold, and the step of calculating the logic synchronization rate is based on cross-correlation analysis of a normalized waveform of the energy consumption time sequence data and the pseudo-random working condition control sequence.
3. The marine ecological environment analysis method based on big data according to claim 1 is characterized by further comprising the step of executing background noise differential elimination based on flow field topology before the step of calculating the cross-correlation coefficient, wherein the step of selecting background reference points which are not affected by emission of controlled objects in a space topological structure of a target sea area based on flow field transmission logic, acquiring background environment data of the background reference points in the same period and correspondingly correcting the background environment data through flow field transmission to generate a background baseline sequence, calculating a differential sequence of environment monitoring time sequence data and the background baseline sequence, and taking the differential sequence as input data for calculating the cross-correlation coefficient to filter regional common mode environment noise.
4. The marine ecological environment analysis method based on big data, which is characterized by comprising the steps of obtaining gridding average flow field data of a target sea area, extracting a streamline path connecting a control object discharge point and an environment monitoring point, carrying out path integration on a flow velocity vector of the gridding average flow field data along the streamline path, and calculating to obtain theoretical transmission time; the method comprises the steps of determining that abnormal fluctuation of environment monitoring time sequence data is generated by a management object and depends on topology consistency check logic, and specifically comprises the steps of determining a verification monitoring point located at the downstream of an environment monitoring point based on geographic information of a target sea area, checking actual monitoring data of the verification monitoring point in secondary lag time, judging whether the actual monitoring data present secondary characteristic response conforming to theoretical attenuation amplitude, if the verification result is negative, determining that the data of the environment monitoring point are abnormal, blocking establishment of responsibility association mapping, and generating a device calibration work order aiming at the environment monitoring point.
5. The marine ecological environment analysis method based on big data according to claim 1, wherein the step of generating the supervision and punishment instruction comprises the steps of obtaining a specific value of local waveform matching degree, generating the supervision and punishment instruction comprising an encryption monitoring requirement and a field check task for a controlled object when the local waveform matching degree is located in a first confidence interval, and generating the supervision and punishment instruction comprising an instant shutdown signal or an administrative punishment data packet for the controlled object when the local waveform matching degree is located in a second confidence interval higher than the first confidence interval.
6. The marine ecological environment analysis method based on big data according to claim 1 is characterized by further comprising the steps of performing defensive detection on data tampering behaviors of the management and control object, specifically comprising the steps of monitoring statistical characteristics of operating time sequence data of the operating condition, responding to variance or information entropy of the operating time sequence data which is lower than a preset threshold value during the non-regular action execution, judging that the management and control object has data modification behaviors, directly generating a field law enforcement instruction with the highest priority, and locking an electronic supervision file of the management and control object.
7. A marine environmental analysis system based on big data, characterized in that it is adapted to implement the method of any of claims 1 to 6, comprising: the data acquisition and space-time mapping module is used for acquiring environment monitoring time sequence data of a target sea area and working condition operation time sequence data of a control object, and determining a transmission lag time window from the control object to an environment monitoring point position based on a preset flow field transmission logic; the cross-correlation calculation module is used for calculating the cross-correlation coefficient of the working condition operation time sequence data and the environment monitoring time sequence data in the transmission delay time window; The initiative attribution right determining module is used for responding to the fact that the cross correlation coefficient falls into a preset gray level verification interval, generating a pseudo-random working condition control sequence, constructing the frequency domain characteristic of the pseudo-random working condition control sequence to be orthogonal with the background main frequency of the current environment monitoring time sequence data, issuing a scheduling instruction containing the pseudo-random working condition control sequence to a management object through an administrative management interface, and executing irregular action according to the pseudo-random working condition control sequence so as to enable the working condition running state of the management object to be correspondingly changed; The fingerprint verification and decision module is used for acquiring verification environment time sequence data after irregular action execution, calculating local waveform matching degree of the verification environment time sequence data and a pseudo-random working condition control sequence, judging that abnormal fluctuation of environment monitoring time sequence data is derived from a control object when the local waveform matching degree exceeds a right-confirming threshold, establishing responsibility association mapping aiming at the control object and generating a supervision punishment instruction, wherein the pseudo-random working condition control sequence is used for demodulating exclusive responsibility subject fingerprints from environment background noise in a multi-source interference environment.

Description

Marine ecological environment analysis method and system based on big data Technical Field The invention relates to a marine ecological environment analysis method and system based on big data, and belongs to the technical field of marine ecological environments. Background The current marine ecological environment administration supervision and management relies on satellite remote sensing, fixed-point buoys and shore base stations to form a three-dimensional monitoring network, water quality environment parameters are collected in real time and are compared with legal environment quality standard numerical values to be used as the basis for judging environment abnormality and triggering administration law enforcement programs, the marine environment has strong dynamic coupling characteristics and high background noise complex system attributes, periodic or random oscillation is generated in environment monitoring indexes due to tidal fluctuation, thermocline fluctuation and seasonal ocean current replacement natural hydrologic processes, background fluctuation is caused by natural factors, the time sequence characteristics and numerical amplitude of the environment monitoring indexes are highly confused with environment abnormality caused by artificial emission, and the existing static judgment logic based on absolute numerical threshold is difficult to strip and determine causal relationship from aliasing monitoring data when facing to unstable hydrologic environments, so that natural fluctuation is misjudged as artificial responsibility, or a specific offensive body cannot be locked under multi-source emission interference. In order to improve supervision attribution accuracy, pollution tracing inversion is usually carried out by increasing sensor deployment density or constructing a high-precision three-dimensional hydrodynamic numerical model, but large-scale application has limitations that high-density hardware array construction and operation and maintenance cost are exponentially increased, nonlinear drifting and invalidation of sensor data are difficult to overcome due to marine organism attachment and seawater corrosion, a complex fluid physical model has the defects of difficult acquisition of boundary conditions, long calculation time consumption and the like, timeliness requirements of administrative supervision cannot be met in terms of real-time blocking and rapid evidence taking aiming at illegal emission behaviors, and under the conditions that multiple emission source spaces are adjacent and emission characteristics are similar, an evidence chain with legal exclusivity is difficult to be constructed only by physical diffusion simulation conclusion, the administrative punishment faces to be held, and the existing application adopts data analysis logic to process attribution blind areas existing in a complex hydrologic background, for example, the Chinese patent application with publication number CN114659503A discloses an artificial intelligence-based marine ecological environment monitoring method, which combines meteorological data and history comparison to identify environment change trend, and establishes data association. However, the method belongs to passive monitoring, relies on statistical correlation of monitoring indexes rather than physical causality, and under the masking of strong periodic background noise, if the emission behavior of a controlled object and the natural hydrologic cycle frequency aliasing or the coexistence of multiple emission sources exist, the passive logic can not distinguish natural fluctuation and artificial emission from the physical level only by observation comparison, and the responsibility main body is difficult to lock. Therefore, how to utilize the existing heterogeneous data resources, reduce the hardware investment scale, decouple the natural background noise and the artificial emission characteristics through the data processing logic, and realize the accurate locking and the real-time supervision of the environment abnormal responsibility main body under the complex hydrologic condition, thereby becoming the technical problem to be solved by the invention. Disclosure of Invention In order to solve the problems in the background technology, the technical scheme of the invention is as follows, a marine ecological environment analysis method based on big data comprises the following steps: acquiring environment monitoring time sequence data of a target sea area and working condition operation time sequence data of a control object, and determining a transmission lag time window from the control object to an environment monitoring point position based on a preset flow field transmission logic; Responding that the cross-correlation coefficient falls into a preset gray verification interval, generating a pseudo-random working condition control sequence, wherein the frequency domain characteristic of the pseudo-random working condition control sequ