Reservoir water ecological health diagnosis method and system

Abstract

The invention provides a reservoir water ecological health diagnosis method and system, and relates to the technical field of reservoir ecological monitoring, and the technical scheme is characterized in that the method comprises the steps: obtaining water quality parameters, biological indexes and environmental data collected by distributed sensors in a reservoir; according to the water quality parameters, the biological indexes and the environmental data, generating reservoir ecological health assessment data through a data fusion algorithm; calculating a reservoir ecological health assessment index according to the reservoir ecological health assessment data; according to the reservoir ecological health assessment index, generating a reservoir ecological health state assessment result; judging whether the ecological state of the reservoir is abnormal or not according to the evaluation result of the ecological health state of the reservoir; and when the ecological state of the reservoir is abnormal, generating an early warning signal. The reservoir water ecological health diagnosis method and system provided by the invention have the advantages that multiple ecological pressure factors are comprehensively considered, and a complex water ecological system is rapidly and accurately evaluated.

Description

Technical Field

[0001] The present application relates to the technical field of reservoir ecological monitoring, and specifically, to a reservoir water ecological health diagnosis method and system. Background Art

[0002] In the ecological health management of large reservoirs, management departments face complex and arduous challenges. Reservoir ecosystems are affected by a variety of factors, including seasonal changes, human activities and natural factors, which makes it difficult for a single monitoring method to fully reflect the ecological health of the reservoir. In order to meet this challenge, management departments have adopted a comprehensive monitoring strategy with multiple points and methods, including water quality parameter detection, biological indicator analysis and remote sensing image analysis.

[0003] However, this comprehensive monitoring strategy has encountered many difficulties in its implementation. First, the geographical characteristics of the reservoir pose challenges to data collection. Due to the vast area and deep water depth of the reservoir, some monitoring equipment has difficulty reaching the deep water area, which affects the comprehensiveness and accuracy of data collection. Secondly, the complexity of the underwater environment further increases the difficulty of data collection. There are still some key deficiencies in the existing reservoir water ecological health diagnosis methods. These methods are difficult to fully consider multiple ecological stress factors such as microplastic pollution, cyanobacteria outbreaks, invasive alien species, and abnormal fish behavior, resulting in an incomplete and inaccurate assessment of the ecological health status of the reservoir. At the same time, the existing methods lack the ability to quickly, comprehensively and accurately assess complex aquatic ecosystems, and it is difficult to strike a balance between ecological security, water supply needs and public concerns.

[0004] In addition, the existing technology does not adequately analyze the structural and functional characteristics of the reservoir ecosystem, making it difficult to accurately predict the changing trends after the project is put into operation. The lack of a reservoir ecological health status assessment technology system based on the structure and function of the ecosystem has cast doubt on the scientificity and reliability of the assessment results.

[0005] In view of the above problems, the existing technology needs to be improved urgently. Summary of the invention

[0006] The purpose of this application is to provide a reservoir water ecological health diagnosis method and system, which aims to solve at least one problem existing in the prior art and has the advantages of comprehensively considering multiple ecological stress factors and quickly and accurately evaluating complex water ecosystems.

[0007] This application provides a reservoir water ecological health diagnosis method, the technical solution is as follows: The method includes: acquiring water quality parameters, biological indicators and environmental data collected by distributed sensors in the reservoir; generating reservoir ecological health assessment data through a data fusion algorithm based on the water quality parameters, biological indicators and environmental data; calculating a reservoir ecological health assessment index based on the reservoir ecological health assessment data; generating a reservoir ecological health status assessment result based on the reservoir ecological health assessment index; judging whether the reservoir ecological status is abnormal based on the reservoir ecological health status assessment result; and generating an early warning signal when the reservoir ecological status is abnormal.

[0008] Furthermore, the present application also proposes that the step of generating reservoir ecological health assessment data through a data fusion algorithm based on the water quality parameters, biological indicators and environmental data includes: obtaining the spatiotemporal distribution information of the water quality parameters, biological indicators and environmental data; performing spatiotemporal interpolation processing on the water quality parameters, biological indicators and environmental data based on the spatiotemporal distribution information; and generating reservoir ecological health assessment data through a multi-source data fusion algorithm based on the data after the spatiotemporal interpolation processing.

[0009] Furthermore, the present application also proposes that the step of performing spatiotemporal interpolation processing on the water quality parameters, biological indicators and environmental data based on the spatiotemporal distribution information includes: obtaining the sampling time interval and sampling point spacing of the water quality parameters, biological indicators and environmental data; calculating the spatiotemporal interpolation coefficients of the water quality parameters, biological indicators and environmental data based on the sampling time interval and sampling point spacing; performing spatiotemporal interpolation calculation on the water quality parameters, biological indicators and environmental data based on the spatiotemporal interpolation coefficients to obtain the water quality parameters, biological indicators and environmental data after interpolation processing.

[0010] Furthermore, the present application also proposes that the step of calculating the spatiotemporal interpolation coefficients of the water quality parameters, biological indicators and environmental data according to the sampling time interval and the sampling point spacing includes: obtaining the historical change trends of the water quality parameters, biological indicators and environmental data; calculating the change rates of the water quality parameters, biological indicators and environmental data at different sampling time intervals according to the historical change trends; calculating the spatial correlation of the water quality parameters, biological indicators and environmental data at different sampling point spacings according to the change rates; generating time interpolation coefficients and spatial interpolation coefficients according to the change rates and the spatial correlations; and combining the time interpolation coefficients and the spatial interpolation coefficients to obtain the spatiotemporal interpolation coefficients.

[0011] Furthermore, the present application also proposes that the step of calculating the change rate of the water quality parameters, biological indicators and environmental data at different sampling time intervals based on the historical change trend includes: obtaining a historical data set of the water quality parameters, biological indicators and environmental data; performing a time series analysis on the historical data set to identify periodic change patterns and long-term change trends; establishing a time series prediction model based on the periodic change patterns and long-term change trends; using the time series prediction model to predict data values ​​at different sampling time intervals; calculating the difference between adjacent predicted data values, and dividing it by the corresponding time interval to obtain the change rate of the water quality parameters, biological indicators and environmental data at different sampling time intervals.

[0012] Furthermore, the present application also proposes that the step of calculating the reservoir ecological health assessment index based on the reservoir ecological health assessment data includes: obtaining water quality parameters, biological indicators and remote sensing data in the reservoir ecological health assessment data; calculating the water quality parameter index based on the water quality parameters, wherein the water quality parameter index includes the weighted sum of dissolved oxygen, pH value, total nitrogen, total phosphorus and temperature; calculating the biological indicator score based on the biological indicators, wherein the biological indicator score includes the weighted sum of algae, fish and plankton; calculating the remote sensing data index based on the remote sensing data, wherein the remote sensing data index includes the weighted sum of the normalized water body index, chlorophyll a and turbidity; calculating the reservoir ecological health assessment index based on the water quality parameter index, the biological indicator score and the remote sensing data index.

[0013] Furthermore, the present application also proposes that the method also includes: obtaining reservoir fish community structure data based on the reservoir ecological health status assessment results; establishing a reservoir fish community structure database based on the reservoir fish community structure data; generating fish release and fishing strategies based on the reservoir fish community structure database; executing aquatic vegetation restoration based on the fish release and fishing strategies; and placing artificial fish nests based on the execution results of the aquatic vegetation restoration.

[0014] Furthermore, the present application also proposes that the step of generating fish release and fishing strategies based on the reservoir fish community structure database includes: obtaining the fish population quantity, age structure and biomass data in the reservoir fish community structure database; calculating the target fish community structure according to a preset reservoir ecological balance model; comparing the target fish community structure with the current fish community structure to obtain difference data; based on the difference data, combined with reservoir water quality parameters and seasonal factors, generating a fish release plan including fish release type, quantity and time; based on the difference data, generating fishing restrictions including fish fishing type, quantity and regional restrictions; integrating the fish release plan and the fishing restrictions to form the fish release and fishing strategies.

[0015] Furthermore, the present application also proposes that the calculation formula for calculating the reservoir ecological health assessment index based on the reservoir ecological health assessment data is: REHM = [Σ(Wi × Pi × Si) × D × T] × E × R in, DREHM is the reservoir ecological health assessment index; Wi is the weight coefficient of each indicator, and its value range is 0-1; Pi is the weight of the monitoring point, ranging from 0 to 1; Si is the seasonal adjustment factor, ranging from 0.8 to 1.2; D is the depth correction coefficient, ranging from 0.7 to 1.0; T is the time response function, T = 1 - e^(-t1 / τ); τ is the characteristic time; t1 is the time interval between the current time point and a certain reference time; E is the ecological pressure function, E = 1 / (1 + e^(-k(MP + BA + IS))); k is the ecological pressure response coefficient; R is the risk adjustment factor, R = min(1, max(0.5, 1 - Σ(δiRFi))); MP is the microplastic pollution index, MP = Σ(μ1C1 + μ2C2 + ... + μnCn); μi is the impact weight coefficient of the i-th type of microplastics, i=1...n; Ci is the concentration of the i-th type of microplastics, i=1...n; BA is the cyanobacterial biomass index, BA = V × N × (1 + sin(2πt2 / 365)); V is the volume of cyanobacteria; N is the number density of cyanobacteria; t2 is the current time point; IS is the impact of invasive species, IS = Σ(ρiPi × Ai); ρi is the ecological impact coefficient of the i-th invasive species; δi is the weight of the i-th risk factor; Ai is the distribution area of ​​the i-th invasive species; RFi is the risk value of the ith risk factor.

[0016] Furthermore, the present application also proposes a reservoir water ecological health diagnosis system, which includes: a data acquisition module, used to obtain water quality parameters, biological indicators and environmental data collected by distributed sensors in the reservoir; a data fusion module, used to generate reservoir ecological health assessment data through a data fusion algorithm based on the water quality parameters, biological indicators and environmental data; an assessment index calculation module, used to calculate the reservoir ecological health assessment index based on the reservoir ecological health assessment data; an assessment result generation module, used to generate a reservoir ecological health status assessment result based on the reservoir ecological health assessment index; a status judgment module, used to judge whether the reservoir ecological status is abnormal based on the reservoir ecological health status assessment result; an early warning module, used to generate an early warning signal when the reservoir ecological status is abnormal.

[0017] From the above, it can be seen that the reservoir water ecological health diagnosis method and system provided in this application realizes a comprehensive and accurate assessment of the ecological health status of the reservoir by adopting data fusion and comprehensive evaluation index calculation technology. It has the advantages of comprehensively considering multiple ecological stress factors and quickly and accurately assessing complex water ecosystems. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of a reservoir water ecological health diagnosis method provided for this application.

[0019] Figure 2 A schematic diagram of the structure of a reservoir water ecological health diagnostic system provided in this application.

[0020] In the figure: 210, data acquisition module; 220, data fusion module; 230, evaluation index calculation module; 240, evaluation result generation module; 250, state judgment module; 260, early warning module. DETAILED DESCRIPTION

[0021] The technical solutions in the present application will be clearly and completely described below in conjunction with the drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0022] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0023] In the ecological health monitoring of large reservoirs, existing technologies face multiple challenges. First, the reservoir area is vast, the water quality is complex and changeable, and it is difficult for a single monitoring method to fully reflect the ecological health of the reservoir due to seasonal changes, human activities and natural factors. Second, the reservoir is deep, some monitoring equipment is difficult to reach the deep water area, and the underwater environment is complex, which affects the accuracy of data collection.

[0024] If the above technical problems cannot be effectively solved, the performance and reliability of the reservoir ecological health monitoring system will be greatly reduced. The incompleteness and inaccuracy of monitoring data will lead to deviations in ecological health assessment results, which may miss important ecological problems or generate false alarms. Data transmission delays will reduce the system's real-time response capabilities, making it impossible to detect and deal with sudden water pollution or ecological disasters in a timely manner. Due to the inability to fully consider multiple ecological stress factors, the system is difficult to accurately assess the structural and functional characteristics of the reservoir ecosystem, which in turn affects the prediction of change trends after the project is put into operation.

[0025] This application proposes a comprehensive reservoir water ecological health diagnosis method to address the multiple challenges faced in monitoring the ecological health of large reservoirs. This method aims to solve problems such as the vast area of ​​reservoirs, complex and changeable water quality conditions, difficulty in monitoring deep water areas, and limited data transmission, while improving the ability to identify and warn of sudden ecological problems.

[0026] Specifically, this application first considers how to comprehensively collect data related to the ecological health of the reservoir. By deploying a distributed sensor network in the reservoir, comprehensive collection of water quality parameters, biological indicators and environmental data can be achieved. These sensors can cover different areas and depths of the reservoir, including deep water areas that are difficult to reach, thus solving the blind spot problem of data collection.

[0027] However, simply collecting data is not enough. How to effectively process and utilize these data becomes the next problem to be solved. This application proposes the use of data fusion algorithms to process multi-source heterogeneous data. This method can integrate different types of data (such as water quality parameters, biological indicators, and environmental data) to generate more comprehensive and reliable reservoir ecological health assessment data.

[0028] Next, this application considers how to convert these assessment data into quantifiable indicators. By calculating the reservoir ecological health assessment index, the complex ecological situation can be simplified into numerical values ​​that are easy to understand and compare. The calculation of this index takes into account multiple factors, including water quality, biodiversity, environmental conditions, etc., thus providing a comprehensive ecological health assessment.

[0029] Based on this evaluation index, this application further proposes a method for generating reservoir ecological health status evaluation results. This result not only includes a quantitative index, but also a qualitative description of the reservoir's ecological status, allowing managers to more intuitively understand the reservoir's ecological health status.

[0030] In order to detect and respond to ecological problems in a timely manner, this application designs a judgment mechanism to determine whether the ecological status of the reservoir is abnormal. This mechanism is based on preset thresholds or historical data comparisons to quickly identify potential ecological risks.

[0031] When an abnormal ecological state is detected, the present application proposes a method for generating an early warning signal. This early warning mechanism can promptly notify management personnel so that they can quickly take appropriate measures to prevent the ecological problem from further deteriorating.

[0032] Therefore, refer to Figure 1 , this application proposes a reservoir water ecological health diagnosis method, the method comprising: S110, obtaining water quality parameters, biological indicators and environmental data collected by distributed sensors in the reservoir; S120, generating reservoir ecological health assessment data through a data fusion algorithm according to the water quality parameters, biological indicators and environmental data; S130, calculating a reservoir ecological health assessment index according to the reservoir ecological health assessment data; S140, generating a reservoir ecological health status assessment result according to the reservoir ecological health assessment index; S150, judging whether the ecological status of the reservoir is abnormal according to the evaluation result of the ecological health status of the reservoir; S160. When the ecological status of the reservoir is abnormal, an early warning signal is generated.

[0033] Among them, distributed sensors refer to multiple sensing devices distributed at different locations in the reservoir, which can be implemented by water quality monitoring buoys, underwater robots or fixed monitoring stations.

[0034] Among them, the data fusion algorithm refers to the computational method of integrating and analyzing multi-source heterogeneous data, which can be implemented using algorithms such as Kalman filtering, Bayesian inference or artificial neural networks.

[0035] Among them, the reservoir ecological health assessment index refers to a comprehensive indicator that quantitatively represents the health status of the reservoir ecosystem, which can be calculated using methods such as weighted summation or fuzzy comprehensive evaluation.

[0036] Among them, the early warning signal refers to the alarm information issued when the ecological status of the reservoir is detected to be abnormal, which can be implemented in the form of SMS notification, sound and light alarm or system pop-up window.

[0037] The core innovation of this application is to propose a comprehensive, real-time and intelligent reservoir water ecological health diagnosis method. This method realizes comprehensive data collection through a distributed sensor network, improves data reliability by using data fusion algorithms, realizes intelligent diagnosis through evaluation index calculation and state judgment, and finally forms an early warning mechanism, which effectively solves the technical problems in large-scale reservoir ecological health monitoring.

[0038] The working principle of this application can be described in detail as follows: First, a distributed sensor network is deployed in the reservoir. These sensors include water quality monitoring buoys, underwater robots, and fixed monitoring stations. Water quality monitoring buoys can float on the water surface and monitor surface water quality parameters in real time. Underwater robots can cruise autonomously and collect data at different water depths. Fixed monitoring stations are installed at key locations in the reservoir to continuously monitor ecological indicators in specific areas. This multi-level sensor layout ensures the comprehensiveness and accuracy of data collection.

[0039] Next, the data collected by all sensors are sent to the data processing center in real time through wireless transmission technology. In the data processing center, data fusion algorithms are used to integrate and analyze multi-source heterogeneous data. Specifically, the Kalman filter algorithm can be used to optimize the fusion of data from different sensors to eliminate noise and redundant information. At the same time, the Bayesian inference method is used to evaluate the reliability of the data and eliminate outliers. In addition, artificial neural network algorithms can be applied to extract key features from massive data to generate more comprehensive and reliable reservoir ecological health assessment data.

[0040] Based on the fused assessment data, the system calculates the reservoir ecological health assessment index. This step can use the weighted summation method to assign different weights to different ecological indicators according to their importance, and comprehensively calculate a quantitative assessment index. At the same time, considering the complexity of the ecosystem, the fuzzy comprehensive evaluation method can also be introduced to better deal with the nonlinear relationship and uncertainty between indicators.

[0041] The system automatically generates the reservoir ecological health status assessment results based on the calculated assessment index and the preset health status classification standards. This result not only includes the quantitative index value, but also includes the qualitative description of the reservoir ecological status, such as "healthy", "sub-healthy" or "unhealthy".

[0042] The system will then compare the current assessment results with historical data and preset thresholds to determine whether the reservoir's ecological status is abnormal. If the assessment index is lower than the warning threshold, or if some key indicators fluctuate abnormally, the system will determine that the ecological status is abnormal.

[0043] Finally, when an abnormal ecological state is detected, the system immediately generates an early warning signal. These signals can be sent through multiple channels, including SMS notifications to managers, sound and light alarms at the monitoring center, and pop-up warning windows on the management system interface. This multiple early warning mechanism ensures that managers can be informed of abnormal situations in a timely manner and take appropriate countermeasures quickly.

[0044] As a preferred implementation, the present application can be implemented in a large reservoir, where 100 water quality monitoring buoys are arranged evenly distributed on the surface of the reservoir. Each buoy is equipped with sensors such as dissolved oxygen, pH value, temperature, turbidity, etc., and the sampling frequency is once every 10 minutes. At the same time, 10 underwater robots are deployed, each equipped with sonar, camera and multi-parameter water quality sensor, which can cruise autonomously in the water depth range of 0-100 meters to collect water quality, biological and environmental data. In addition, 20 fixed monitoring stations are installed at key locations around the reservoir to monitor the water quality of the rivers entering the reservoir, meteorological conditions and the surrounding ecological environment.

[0045] All data collected by sensors are transmitted to the data processing center in real time. In the data processing center, a data fusion algorithm based on deep learning is used to process multi-source data. This algorithm can effectively handle data inconsistencies between different sensors and extract key features from massive data. After fusion processing, the system generates reservoir ecological health assessment data once an hour.

[0046] Based on the assessment data, the system uses an improved fuzzy comprehensive evaluation method to calculate the reservoir ecological health assessment index. This method takes into account multiple key ecological indicators, including dissolved oxygen, total nitrogen, total phosphorus, chlorophyll a, plankton biodiversity index, etc., and assigns different weights to each indicator based on expert opinions. The calculated assessment index ranges from 0 to 100, where 0-60 is unhealthy, 60-80 is sub-healthy, and 80-100 is healthy.

[0047] The system generates a reservoir ecological health status assessment report every day, including the assessment index, detailed data of various indicators and a qualitative description of the ecological status. At the same time, the system monitors the changes in the assessment index in real time. If the index is lower than 60 or drops by more than 10% for three consecutive days, the system will determine that the ecological status is abnormal.

[0048] Once an abnormality is detected, the system immediately notifies managers via text messages and phone calls, displays warning messages on the large screen at the monitoring center, and automatically generates emergency plan suggestions. This timely early warning mechanism enables managers to quickly respond to potential ecological risks and effectively prevent water quality deterioration and ecological disasters.

[0049] In some of the above-mentioned embodiments, during the implementation of the present application, there are still problems such as incomplete data collection and low data processing efficiency.

[0050] In this regard, the present application further proposes to obtain the spatiotemporal distribution information of water quality parameters, biological indicators and environmental data; perform spatiotemporal interpolation processing on water quality parameters, biological indicators and environmental data based on the spatiotemporal distribution information; and generate reservoir ecological health assessment data through a multi-source data fusion algorithm based on the data after spatiotemporal interpolation processing.

[0051] This application effectively solves the problem of incomplete data collection due to the vast reservoir area and limited monitoring points by introducing spatiotemporal distribution information and spatiotemporal interpolation processing. At the same time, the multi-source data fusion algorithm is used to improve data processing efficiency and provide more comprehensive and accurate data support for reservoir ecological health assessment.

[0052] Specifically, the application first obtains the spatiotemporal distribution information of water quality parameters, biological indicators and environmental data. This information may include the geographical coordinates of each monitoring point, sampling time, sampling depth, etc. The acquisition method can be to record the location of the monitoring point through the GPS positioning system, or to extract relevant information using the reservoir geographic information system (GIS) database.

[0053] Next, based on the acquired spatiotemporal distribution information, water quality parameters, biological indicators and environmental data are subjected to spatiotemporal interpolation. The purpose of spatiotemporal interpolation is to estimate data values ​​at unsampled time points and spatial locations, thereby obtaining a more continuous and complete data set. Commonly used spatiotemporal interpolation methods include Kriging, Inverse Distance Weighted (IDW), etc. For example, the spatiotemporal Kriging interpolation method can be used, which takes into account both the spatial correlation and temporal correlation of the data, and can better capture the spatiotemporal variation characteristics of water quality parameters, biological indicators and environmental data.

[0054] When performing spatiotemporal interpolation, appropriate interpolation methods can be selected according to the characteristics of different parameters. For parameters that change relatively slowly (such as water temperature), simple linear interpolation can be used; for parameters with obvious spatial heterogeneity (such as dissolved oxygen), more complex geostatistical methods may be required. During the interpolation process, factors such as the topographic characteristics of the reservoir and the direction of water flow need to be considered to improve the accuracy of the interpolation results.

[0055] After the spatiotemporal interpolation processing is completed, a data set covering the entire reservoir with a higher temporal resolution is obtained. This data set contains the original monitoring data and the estimated data obtained through interpolation. In this way, the discontinuity of the original data in time and space is overcome, providing a better foundation for subsequent data fusion and analysis.

[0056] Finally, based on the data processed by spatiotemporal interpolation, the reservoir ecological health assessment data is generated through a multi-source data fusion algorithm. The purpose of the multi-source data fusion algorithm is to integrate data from different sources and scales to obtain more comprehensive and accurate information. Commonly used multi-source data fusion algorithms include Bayesian fusion, Kalman filtering, and evidence theory fusion.

[0057] In the specific implementation, a multi-source data fusion algorithm based on evidence theory can be used. First, water quality parameters, biological indicators and environmental data are used as different sources of evidence. Then, according to the reliability and importance of each source of evidence, appropriate weights are assigned to them. Next, the combination rules of Dempster-Shafer evidence theory are used to fuse the information of each source of evidence. Finally, the fused reservoir ecological health assessment data is obtained through decision rules.

[0058] For example, suppose there are three main sources of evidence: water quality parameters (such as dissolved oxygen, pH), biological indicators (such as algae quantity, fish diversity index) and environmental data (such as water temperature, light intensity). Each source of evidence has been processed by spatiotemporal interpolation to form a continuous data field covering the entire reservoir. Through the evidence theory fusion algorithm, a comprehensive reservoir ecological health assessment index can be obtained. This index not only reflects the water quality status, but also includes the influence of biological community structure and environmental factors, thereby more comprehensively characterizing the ecological health status of the reservoir.

[0059] By adopting the above-mentioned technical solutions, the present application can effectively solve the problems existing in the prior art such as incomplete data collection and low data processing efficiency. Specifically, the present application fills the gaps in time and space of the original monitoring data through spatiotemporal interpolation processing, making the reservoir ecological health assessment data more continuous and complete. The multi-source data fusion algorithm comprehensively considers the characteristics and reliability of different types of data, reducing the errors and uncertainties that may be caused by a single data source. The use of interpolation and fusion algorithms can quickly process a large amount of raw data and generate high-quality assessment data to meet the needs of real-time monitoring and early warning. By integrating water quality parameters, biological indicators and environmental data, the assessment results obtained more comprehensively reflect the health status of the reservoir ecosystem.

[0060] Compared with the prior art, the innovations of this application are mainly reflected in the following aspects: First, existing technologies usually focus on a single type of data, such as using only water quality parameters for evaluation. This application introduces spatiotemporal distribution information and multi-source data fusion to achieve the comprehensive utilization of water quality parameters, biological indicators and environmental data, greatly improving the comprehensiveness and accuracy of the evaluation.

[0061] Secondly, traditional methods often use simple interpolation techniques or directly use discrete monitoring data, which is difficult to accurately reflect the dynamic changes of reservoir ecosystems. This application introduces spatiotemporal interpolation processing, which not only improves the spatiotemporal resolution of data, but also better captures the spatiotemporal variation characteristics of water quality parameters and biological indicators.

[0062] Finally, existing data fusion methods are mostly limited to the fusion of similar data, such as the fusion of multiple water quality parameters. The multi-source data fusion algorithm used in this application can process heterogeneous data and organically combine different types of data such as water quality, biology and environment, so as to obtain more comprehensive and reliable evaluation results.

[0063] In some specific implementations, the spatiotemporal interpolation process may interpolate time and space separately, that is, first interpolate in the time dimension to supplement the missing time point data, and then interpolate in the space dimension to fill in the unknown points of the spatial distribution.

[0064] In some other specific implementations, time and space are jointly interpolated, that is, both time and space dimensions are considered simultaneously to construct a joint time-space model, and the relative weights of time and space are dynamically adjusted through the joint time-space interpolation coefficients.

[0065] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to accurately process the spatiotemporal distribution information of water quality parameters, biological indicators and environmental data to improve the accuracy and reliability of reservoir ecological health assessment.

[0066] In this regard, the present application further proposes sampling time intervals and sampling point spacings for obtaining water quality parameters, biological indicators, and environmental data; based on the sampling time intervals and sampling point spacings, the spatiotemporal interpolation coefficients of water quality parameters, biological indicators, and environmental data are calculated; based on the spatiotemporal interpolation coefficients, spatiotemporal interpolation calculations are performed on the water quality parameters, biological indicators, and environmental data to obtain the water quality parameters, biological indicators, and environmental data after interpolation processing.

[0067] Among them, the spatiotemporal interpolation coefficient is a parameter used to quantify and optimize the data interpolation weight in the spatiotemporal dimension. It combines the rate of change of the time dimension and the correlation of the spatial dimension to dynamically adjust the weight distribution of the interpolation model in time and space, thereby achieving more accurate interpolation calculations.

[0068] The reservoir water ecological health diagnosis method proposed in this application aims to solve the problems of large reservoir area, complex and changeable water quality conditions, and difficult to reach deep water areas. By performing spatiotemporal interpolation processing on water quality parameters, biological indicators and environmental data, a comprehensive assessment of the ecological health status of the reservoir can be achieved.

[0069] Specifically, the method of the present application first obtains the sampling time intervals and sampling point spacings of water quality parameters, biological indicators and environmental data. These data come from multiple monitoring points distributed in different areas of the reservoir, including sampling points at different depths such as the water surface, the middle and bottom of the water body. The sampling time interval may vary depending on different parameters. For example, water temperature may be sampled once an hour, while parameters such as total nitrogen and total phosphorus may be sampled once a day. The spacing between sampling points depends on the distribution density of the monitoring points, which may range from tens of meters to hundreds of meters.

[0070] Next, the spatiotemporal interpolation coefficients of water quality parameters, biological indicators and environmental data are calculated based on the sampling time interval and the spacing between sampling points. The calculation of spatiotemporal interpolation coefficients takes into account the temporal and spatial variation characteristics of the data. For example, for parameters that change faster (such as dissolved oxygen), a smaller temporal interpolation coefficient can be used; while for parameters that change more slowly (such as total phosphorus), a larger temporal interpolation coefficient can be used. Similarly, for parameters with a more uniform spatial distribution, a larger spatial interpolation coefficient can be used; while for parameters with an uneven spatial distribution, a smaller spatial interpolation coefficient is required. The speed of change and whether the space is uniform can be judged based on preset rules.

[0071] The process of calculating the spatiotemporal interpolation coefficients can use methods such as Kriging interpolation, inverse distance weighted method, or spline interpolation, which can be selected and optimized according to the characteristics and distribution of the data. For example, for parameters with obvious spatial correlation, Kriging interpolation can be used; while for parameters with more random spatial distribution, inverse distance weighted method is more suitable.

[0072] Finally, according to the calculated spatiotemporal interpolation coefficients, the water quality parameters, biological indicators and environmental data are interpolated in time and space to obtain the interpolated water quality parameters, biological indicators and environmental data. This step realizes the expansion and supplement of the original sampling data, and can obtain the parameter estimation values ​​at any time and any location in the reservoir.

[0073] For an unsampled location, the data of surrounding known sampling points can be used in combination with spatial interpolation coefficients for weighted average calculation. Similarly, for time points between sampling intervals, the data of adjacent time points can be used in combination with temporal interpolation coefficients for interpolation calculation. These calculations can be implemented using specialized data processing software or custom algorithms.

[0074] Through this method, the present application can effectively solve the data sparsity problem in reservoir monitoring and improve the spatial and temporal resolution of reservoir ecological health assessment. This is of great significance for timely detection of ecological problems in local waters, such as algal blooms or abnormal fish behavior. At the same time, this method also helps to reduce the number of monitoring equipment and reduce monitoring costs, especially in deep water areas or areas with poor communication conditions.

[0075] As a preferred embodiment, the method of the present application can be applied in a large reservoir, specifically with 50 fixed monitoring points distributed in different areas and water depths of the reservoir, and each monitoring point is equipped with a multi-parameter water quality sensor, biological indicator collection equipment and environmental data recorder.

[0076] In this embodiment, water quality parameters (such as dissolved oxygen, pH value, total nitrogen, and total phosphorus) are collected once an hour; biological indicators (such as chlorophyll a concentration and plankton density) are collected once a day; environmental data (such as water temperature and light intensity) are collected every 15 minutes. The spacing between sampling points is set according to the reservoir topography and water quality variation characteristics, ranging from 50 meters to 500 meters.

[0077] First, obtain the information of these sampling time intervals and sampling point spacing. Then, use geostatistical methods (such as variogram analysis) to calculate the spatial correlation of different parameters, and combine time series analysis methods (such as autocorrelation function analysis) to determine the temporal correlation. Based on these analysis results, calculate the corresponding spatiotemporal interpolation coefficients for each parameter.

[0078] For example, for a parameter that changes rapidly, such as dissolved oxygen, the temporal interpolation coefficient may be 0.8 and the spatial interpolation coefficient may be 0.6; while for a parameter that changes slowly, such as total phosphorus, the temporal interpolation coefficient may be 0.95 and the spatial interpolation coefficient may be 0.85.

[0079] Next, these interpolation coefficients are used to interpolate the original data. For example, for a location in the reservoir where no monitoring point is set, the data of the surrounding known monitoring points can be used in combination with the spatial interpolation coefficients to perform Kriging interpolation to obtain the parameter estimate for that location. Similarly, for the time point between two samplings, the linear interpolation or spline interpolation method can be used in combination with the time interpolation coefficients to calculate the parameter estimate for that time point.

[0080] Through this method, the originally sparse data of 50 monitoring points can be expanded into a continuous data set covering the entire reservoir with a temporal resolution of hours. This greatly improves the accuracy and spatiotemporal resolution of reservoir ecological health assessment, allowing management departments to grasp the ecological health status of the reservoir more comprehensively and timely.

[0081] Compared with the existing technology, the method of this application fully considers the spatiotemporal variation characteristics of water quality parameters, biological indicators and environmental data, and improves the accuracy of the interpolation results by calculating personalized spatiotemporal interpolation coefficients. Secondly, it can effectively deal with the problems of uneven distribution of sampling points and inconsistent sampling frequencies, and adapt to the complex actual situation of reservoir monitoring. Finally, through interpolation processing, this method significantly improves the spatiotemporal resolution of reservoir ecological health assessment, providing the possibility for timely discovery and early warning of ecological problems.

[0082] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to accurately calculate the spatiotemporal interpolation coefficients to improve the accuracy of the reservoir ecological health assessment data.

[0083] In this regard, the present application further proposes to calculate the spatiotemporal interpolation coefficients of water quality parameters, biological indicators and environmental data after obtaining the sampling time intervals and sampling point spacings of these data. Specifically, first obtain the historical change trends of water quality parameters, biological indicators and environmental data, and then calculate the change rates of these data at different sampling time intervals based on the historical change trends. Next, calculate the spatial correlation of these data at different sampling point spacings based on the change rate. Finally, generate the time interpolation coefficient and the space interpolation coefficient based on the change rate and spatial correlation, and combine these two coefficients to obtain the final spatiotemporal interpolation coefficient.

[0084] This method of the present application can more accurately calculate the spatiotemporal interpolation coefficients by considering the historical change trend, time change rate and spatial correlation. This method not only considers the temporal change characteristics of the data, but also the spatial distribution characteristics, so as to more comprehensively reflect the dynamic changes of the reservoir ecosystem.

[0085] Specifically, the method of the present application first obtains the historical change trend. The historical change trend can come from long-term monitoring records, including data in different seasons and under different climatic conditions. By analyzing these data, the historical change trend of the data can be identified. For example, some parameters may show obvious seasonal fluctuations, while other parameters may have a trend of increasing or decreasing year by year.

[0086] Based on the identified historical change trends, the data values ​​at different sampling time intervals are predicted. For example, hourly, daily, weekly or monthly data values ​​can be predicted. Then, the change rates of water quality parameters, biological indicators and environmental data at different sampling time intervals are calculated.

[0087] The advantage of this method is that it can capture both short-term fluctuations and long-term trends in the data, thus more accurately reflecting the dynamic changes in the reservoir ecosystem. For example, for parameters that are significantly affected by seasons, such as water temperature or dissolved oxygen, this method can accurately predict their rate of change in different seasons. For parameters affected by human activities, such as total nitrogen or total phosphorus, this method can reflect their long-term changing trends.

[0088] Furthermore, the present application calculates the spatial correlation of water quality parameters, biological indicators and environmental data at sampling time intervals based on the calculated change rates.

[0089] In some embodiments, the smaller the difference in the rate of change between two sampling points is, the higher the spatial correlation is.

[0090] In some other embodiments, the spatial correlation is corrected by adjusting the distance weight, taking into account the spatial distance and the difference in the rate of change of the sampling points.

[0091] This application uses information from the time dimension to infer relationships in the spatial dimension. Specifically, for each sampling point, the rate of change of its parameters (such as water temperature, dissolved oxygen, etc.) over time is calculated. This rate of change may be hourly, daily, or weekly. For each pair of sampling points in the reservoir, their rates of change are compared.

[0092] If the rates of change of two points are very close, they are considered to have high spatial correlation.

[0093] If the rates of change differ widely, they are considered to have low spatial correlation.

[0094] For example, when observing the algae concentration in the reservoir, the algae concentrations at points A and B are increasing rapidly (similar high rates of change), while the algae concentration at point C remains stable (low rate of change).

[0095] According to this method, it can be inferred that: There may be a high spatial correlation between points A and B (they may be under similar environmental conditions); The spatial correlation between A / B and C is low (probably under different environmental conditions).

[0096] The innovation of this method is that it does not rely on static spatial distances, but instead uses the dynamic nature of the data to infer spatial relationships. This is particularly valuable for complex and dynamic ecosystems such as reservoirs, and can capture subtle relationships that traditional spatial analysis methods may overlook.

[0097] The above method is suitable for data where the temporal variation pattern is directly related to the spatial correlation.

[0098] In addition, for cases where temporal variation patterns are not directly related to spatial correlation, spatial autocorrelation analysis methods can be used to calculate the spatial correlation between different sampling points. This analysis can reveal the ecological connections between different areas in the reservoir and help to perform more accurate spatial interpolation.

[0099] Finally, the present application generates temporal interpolation coefficients and spatial interpolation coefficients according to the calculated rate of change and spatial correlation.

[0100] Specifically, in some specific implementations, the change rate may be normalized, and the time interpolation coefficient is proportional to the change rate. The greater the change rate, the greater the time interpolation coefficient.

[0101] Specifically, in some specific implementations, the spatial correlation may be normalized, and the spatial interpolation coefficient is proportional to the correlation. The higher the correlation, the larger the spatial interpolation coefficient.

[0102] The temporal interpolation coefficient reflects the changing characteristics of the data in the temporal dimension, while the spatial interpolation coefficient reflects the distribution characteristics of the data in the spatial dimension. The two coefficients are combined to obtain the final spatiotemporal interpolation coefficient.

[0103] This combination operation can be performed in a weighted average manner, namely: Spatiotemporal interpolation coefficient = w1 * time interpolation coefficient + w2 * space interpolation coefficient; Among them, w1 and w2 are weight coefficients, which can be adjusted according to specific application scenarios. For example, for parameters that change rapidly, the weight of the time interpolation coefficient can be increased; for parameters with large spatial distribution differences, the weight of the spatial interpolation coefficient can be increased.

[0104] As a preferred implementation, the following specific embodiments may be considered: Assume that in a large reservoir, it is necessary to perform spatiotemporal interpolation of dissolved oxygen (DO). First, historical DO data for the past five years are collected, including daily measurements at 10 monitoring points. Through time series analysis, it is found that DO shows obvious seasonal changes, with lower values ​​in summer and higher values ​​in winter.

[0105] Use the SARIMA model to perform time series forecasts on DO and get hourly, daily, and weekly forecasts. Calculate the difference between adjacent forecasts to get the rate of change at different time intervals. For example, the daily rate of change of DO may be 0.5 mg / L / day in summer, while it may be 0.2 mg / L / day in winter.

[0106] Next, the spatial correlation between the 10 monitoring points was calculated using the Moran's index. It was assumed that the monitoring points within 1 km of each other had a high spatial correlation (Moran's index > 0.8), while the monitoring points more than 5 km apart had a low spatial correlation (Moran's index < 0.3).

[0107] Based on these results, the temporal interpolation coefficient and the spatial interpolation coefficient are generated. For example, for DO interpolation in summer, the temporal interpolation coefficient can be set to 0.7 and the spatial interpolation coefficient can be set to 0.3; while for DO interpolation in winter, the temporal interpolation coefficient can be set to 0.4 and the spatial interpolation coefficient can be set to 0.6.

[0108] Finally, the spatiotemporal interpolation coefficients are obtained through weighted averaging.

[0109] Compared with traditional interpolation methods, the method of this application takes into account the temporal variation characteristics of the data and can more accurately reflect the dynamic changes of water quality parameters, biological indicators and environmental data. Taking into account the spatial distribution characteristics of the data, it can better reflect the ecological connection between different areas of the reservoir. By combining the temporal and spatial interpolation coefficients, the comprehensive consideration of the temporal and spatial dimensions is achieved, and the accuracy of the interpolation results is improved. The interpolation strategy can be flexibly adjusted according to the characteristics of different parameters and seasonal changes, and it has strong adaptability. It can better handle missing values ​​and outliers in monitoring data, and improve the reliability of reservoir ecological health assessment.

[0110] Through this precise spatiotemporal interpolation coefficient calculation method, this application can more accurately generate reservoir ecological health assessment data, thereby providing more reliable decision-making support for reservoir management departments and effectively improving the accuracy and practicality of reservoir water ecological health diagnosis.

[0111] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to accurately calculate the change rate of water quality parameters, biological indicators and environmental data at different sampling time intervals.

[0112] In this regard, the present application further proposes to obtain historical data sets of water quality parameters, biological indicators and environmental data; perform time series analysis on the historical data sets to identify periodic change patterns and long-term change trends; establish a time series prediction model based on the periodic change patterns and long-term change trends; use the time series prediction model to predict data values ​​at different sampling time intervals; calculate the difference between adjacent predicted data values ​​and divide them by the corresponding time interval to obtain the change rate of water quality parameters, biological indicators and environmental data at different sampling time intervals.

[0113] The reservoir water ecological health diagnosis method proposed in this application can more accurately predict and calculate the data change rate at different sampling time intervals by deeply analyzing and modeling historical data. This method takes into account the historical change trends of water quality parameters, biological indicators and environmental data, including periodic changes and long-term trends, thereby improving the accuracy and reliability of the change rate calculation.

[0114] Specifically, the technical solution of the present application first obtains historical data sets, which contain long-term observation records of water quality parameters, biological indicators and environmental data. By performing time series analysis on these historical data, the periodic change patterns and long-term change trends of the data can be identified. For example, some water quality parameters may show seasonal fluctuations, while some biological indicators may have long-term growth or decline trends.

[0115] Based on the identified periodic change patterns and long-term change trends, a time series forecasting model is established. This model can be an autoregressive integrated moving average model (ARIMA), an exponential smoothing model, or other suitable time series forecasting model. The choice of model depends on the specific characteristics and change patterns of the data.

[0116] The established time series prediction model is used to predict the data values ​​at different sampling time intervals. The sampling time intervals here can be different scales such as hours, days, weeks or months, which are determined according to the actual monitoring needs. Through the prediction model, the data prediction values ​​at these different time intervals can be obtained.

[0117] Finally, the difference between adjacent predicted data values ​​is calculated and divided by the corresponding time interval to obtain the change rate of water quality parameters, biological indicators and environmental data at different sampling time intervals. This method takes into account the historical change law of the data and can more accurately reflect the speed of change of the data at different time scales.

[0118] Furthermore, the technical solution of the present application can be implemented by the following steps: First, historical records of water quality parameters (such as dissolved oxygen, pH value, total nitrogen, total phosphorus, etc.), biological indicators (such as algae density, fish diversity index, etc.), and environmental data (such as water temperature, light intensity, etc.) over the past few years were extracted from the database of the reservoir monitoring system to form a complete historical data set.

[0119] Secondly, use time series analysis tools, such as the forecast package in R or the statsmodels library in Python, to analyze historical data sets. Through seasonal decomposition and trend analysis, the periodic change patterns (such as seasonal fluctuations) and long-term change trends (such as linear growth or exponential decay) of the data can be identified.

[0120] Then, according to the identified change patterns and trends, the appropriate time series forecasting model is selected. For example, for water temperature data with obvious seasonality, a seasonal ARIMA model can be used; for total nitrogen data showing a long-term growth trend, an exponential smoothing model with a trend term can be used.

[0121] Next, the selected forecast model is used to forecast the data values ​​at different sampling time intervals (such as hourly, daily, and weekly). The forecast time range can be set to the next year to cover the complete seasonal cycle.

[0122] Finally, calculate the difference between adjacent forecast data values ​​and divide it by the corresponding time interval. For example, for daily forecast values, calculate the difference between two consecutive forecast values ​​and divide it by 1 day; for weekly forecast values, calculate the difference between two consecutive forecast values ​​and divide it by 7 days. In this way, we can get the rate of change at different sampling time intervals.

[0123] As a preferred implementation, the following specific examples may be considered: Suppose we are concerned about the dissolved oxygen content in the reservoir. First, we extract the daily dissolved oxygen data for the past five years from the historical data. Through time series analysis, we find that the dissolved oxygen content has obvious seasonal fluctuations, which is lower in summer and higher in winter, and has a slight annual decreasing trend.

[0124] Based on these characteristics, the seasonal ARIMA model is selected for modeling. The model is used to make daily, weekly, and monthly forecasts of dissolved oxygen levels for the next year. For example, the forecast results may show: Daily change rate: an average of ±0.05 mg / L / day, with larger fluctuations in summer, up to ±0.1 mg / L / day.

[0125] Weekly rate of change: average ±0.2 mg / L / week, with gentler changes in spring and autumn.

[0126] Monthly rate of change: Monthly changes between winter and summer can reach 1.5 mg / L / month.

[0127] These rate of change data provide an important basis for subsequent spatiotemporal interpolation and data fusion, and help to more accurately assess the ecological health of the reservoir.

[0128] By adopting this method based on historical data analysis and time series prediction, the present application can more accurately calculate the rate of change of water quality parameters, biological indicators and environmental data at different sampling time intervals. This method takes into account the historical change law and periodic characteristics of the data, and can better reflect the dynamic changes of the reservoir ecosystem than simple linear interpolation or average calculation.

[0129] The method of the present application can more accurately estimate the values ​​of missing data points when performing spatiotemporal interpolation by taking into account historical change trends and periodic patterns. Since the prediction model takes into account the changing patterns of historical data, potential ecological anomalies can be identified more quickly when there is a significant deviation between the actual monitoring data and the predicted values. By analyzing the rate of change at different time intervals, it can provide a basis for formulating more efficient sampling plans, thereby reducing monitoring costs while ensuring data quality. More accurate rate of change calculations help generate more reliable reservoir ecological health assessment data, thereby improving the accuracy of the overall assessment results.

[0130] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to accurately calculate the reservoir ecological health assessment index.

[0131] In this regard, the present application further proposes to obtain water quality parameters, biological indicators and remote sensing data in the reservoir ecological health assessment data; calculate the water quality parameter index based on the water quality parameters, wherein the water quality parameter index includes the weighted sum of dissolved oxygen, pH value, total nitrogen, total phosphorus and temperature; calculate the biological indicator score based on the biological indicators, wherein the biological indicator score includes the weighted sum of algae, fish and plankton; calculate the remote sensing data index based on the remote sensing data, wherein the remote sensing data index includes the weighted sum of the normalized water body index, chlorophyll a and turbidity; calculate the reservoir ecological health assessment index based on the water quality parameter index, the biological indicator score and the remote sensing data index.

[0132] The reservoir water ecological health diagnosis method proposed in this application can more comprehensively and accurately evaluate the ecological health status of the reservoir by comprehensively considering water quality parameters, biological indicators and remote sensing data. The method first obtains water quality parameters, biological indicators and remote sensing data in the reservoir ecological health assessment data, which are derived from the collection of distributed sensors and the processing of data fusion algorithms.

[0133] Specifically, the calculation of the water quality parameter index includes the weighted sum of dissolved oxygen, pH value, total nitrogen, total phosphorus and temperature. These parameters can reflect the basic physical and chemical properties and nutritional status of the water body. For example, dissolved oxygen reflects the self-purification ability of the water body, pH value reflects the acid-base balance of the water body, total nitrogen and total phosphorus reflect the degree of eutrophication of the water body, and temperature affects the survival and reproduction of aquatic organisms.

[0134] The calculation of biological indicator scores includes the weighted sum of algae, fish and plankton. These biological indicators can directly reflect the health of aquatic ecosystems. The types and quantity of algae can reflect the nutritional status and pollution level of water bodies, the diversity and quantity of fish can reflect the integrity of aquatic ecosystems, and plankton is an important part of the aquatic food chain, and its community structure can reflect the ecological balance of water bodies.

[0135] The calculation of remote sensing data index includes the weighted sum of normalized water index, chlorophyll a and turbidity. These indicators can be quickly obtained through remote sensing technology, have a wide coverage, and can reflect the macro characteristics of water bodies. The normalized water index can be used to identify the scope of water bodies and monitor water level changes. Chlorophyll a reflects the biomass of phytoplankton in water bodies, and turbidity reflects the transparency and suspended matter content of water bodies.

[0136] By taking into account the water quality parameter index, biological indicator score and remote sensing data index, this application can calculate a comprehensive reservoir ecological health assessment index. This comprehensive assessment method can better reflect the overall status of the reservoir ecosystem and avoid the one-sidedness that may be caused by a single indicator.

[0137] As a preferred implementation, the water quality parameter index can be calculated using the following formula: WQI = Σ(Wi × Ci) Where WQI is the water quality parameter index, Wi is the weight coefficient of each water quality parameter, and Ci is the standardized value of each water quality parameter. The standardization process can use linear interpolation or piecewise function method to convert parameters of different dimensions into scores of 0-100.

[0138] The biological indicator score can be calculated using the following formula: BIS = Σ(Ai × Bi) Among them, BIS is the biological indicator score, Ai is the weight coefficient of each biological indicator, and Bi is the score of each biological indicator. The score can be based on ecological indicators such as biodiversity index and dominance index.

[0139] The remote sensing data index can be calculated using the following formula: RSI = Σ(Ri × Si) Among them, RSI is the remote sensing data index, Ri is the weight coefficient of each remote sensing indicator, and Si is the standardized value of each remote sensing indicator.

[0140] The final reservoir ecological health assessment index can be calculated by the following formula: REHM = α × WQI + β × BIS + γ × RSI Among them, REHM is the reservoir ecological health assessment index, α, β, and γ are the weight coefficients of water quality parameter index, biological indicator score and remote sensing data index, respectively, and α + β + γ = 1.

[0141] In practical applications, the weight coefficients of various indicators can be adjusted according to the characteristics and management needs of specific reservoirs. For example, for drinking water sources, the weight of water quality parameter index can be appropriately increased; for ecological protection areas, the weight of biological indicator score can be increased; for large reservoirs, the weight of remote sensing data index can be increased.

[0142] Through this method, the present application can more flexibly and accurately evaluate the ecological health status of the reservoir. Compared with the prior art, the method of the present application can comprehensively reflect the status of the reservoir ecosystem by comprehensively considering water quality parameters, biological indicators and remote sensing data. The use of data fusion algorithm to process multi-source data improves the accuracy and reliability of the assessment. By adjusting the weight coefficient, it can adapt to reservoirs of different types and uses. Using distributed sensors and remote sensing technology, real-time monitoring and evaluation of the ecological health status of the reservoir can be achieved. The evaluation index system can be expanded as needed, such as adding emerging indicators such as microplastic pollution and cyanobacteria biomass.

[0143] The method of this application, by comprehensively utilizing multiple data sources and advanced data processing technologies, provides reservoir management departments with a more scientific, comprehensive and flexible ecological health assessment tool, which helps to timely discover reservoir ecological problems and formulate targeted management measures, thereby better protecting the reservoir ecosystem and ensuring the sustainable use of water resources.

[0144] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to carry out effective ecological restoration and management based on the results of the reservoir ecological health status assessment.

[0145] In this regard, the present application further proposes to obtain reservoir fish community structure data based on the reservoir ecological health status assessment results; establish a reservoir fish community structure database based on the reservoir fish community structure data; generate fish release and fishing strategies based on the reservoir fish community structure database; implement aquatic vegetation restoration based on the fish release and fishing strategies; and place artificial fish nests based on the implementation results of aquatic vegetation restoration.

[0146] The method proposed in this application aims to solve the problems in existing technologies, such as the difficulty in comprehensively considering multiple ecological stress factors, the lack of rapid, comprehensive and accurate assessment capabilities for complex aquatic ecosystems, and the difficulty in balancing ecological security, water supply needs and public concerns. By introducing a series of measures such as fish community structure data analysis, generating fish stocking and fishing strategies, implementing aquatic vegetation restoration, and placing artificial fish nests, this application can more comprehensively assess and improve the ecological health of reservoirs.

[0147] The technical solution of this application specifically includes the following steps: First, based on the results of the reservoir ecological health assessment, obtain the reservoir fish community structure data. This step can be achieved in a variety of ways, such as using sonar technology to conduct fish resource surveys or cage sampling methods to conduct fish population analysis. The data obtained may include information such as fish species, quantity, and body length distribution.

[0148] Secondly, a reservoir fish community structure database is established based on the acquired reservoir fish community structure data. This database can use a relational database or a time series database to store and manage data. The database may contain information such as fish population dynamics, age structure, biomass, and the changing trends of these data over time and space.

[0149] Next, based on the established reservoir fish community structure database, fish stocking and fishing strategies are generated. This step can use a multi-objective optimization algorithm to comprehensively consider goals such as water quality improvement, ecological balance maintenance, and sustainable use of fishery resources to generate the optimal fish stocking and fishing plan.

[0150] Then, based on the generated fish stocking and fishing strategies, aquatic vegetation restoration is performed. Aquatic vegetation restoration can include planting various types of aquatic vegetation, including submerged plants, floating leaf plants, and emergent plants. The selection of vegetation and planting density need to be determined according to the specific conditions of the reservoir and the structure of the fish community to ensure that the vegetation can provide a suitable habitat and food source for fish.

[0151] Finally, artificial fish nests are placed according to the results of the aquatic vegetation restoration. The design and placement of artificial fish nests need to take into account the spawning habits and habitat needs of different fish. For example, artificial turf or bionic structures made of environmentally friendly materials can be used as fish nests, and the appropriate placement location can be selected according to the water depth and water flow conditions.

[0152] The results of the analysis of fish community structure data directly affect the formulation of fish stocking and fishing strategies. In turn, fish stocking and fishing strategies affect the aquatic vegetation restoration plan, because different fish have different effects on aquatic vegetation. The restoration of aquatic vegetation will affect the strategy of artificial fish nest placement, because aquatic vegetation can provide natural spawning and habitat for some fish.

[0153] Through this systematic approach, this application can more comprehensively evaluate and improve the ecological health of reservoirs. Specifically, by analyzing the structure of fish communities, abnormal changes in reservoir ecosystems can be discovered in a timely manner, such as a sharp decline in certain fish populations or the invasion of alien species. Fish stocking and fishing strategies based on these data can help maintain the balance of reservoir ecosystems and prevent over-breeding of certain fish or the extinction of endangered fish.

[0154] Restoring aquatic vegetation can not only provide habitats and food sources for fish, but also absorb excess nutrients in the water and improve water quality. The placement of artificial fish nests can provide additional spawning sites for fish and promote the natural reproduction of fish populations. These measures work together to significantly improve the stability and self-repair ability of the reservoir ecosystem.

[0155] In practical applications, the method of the present application can be flexibly adjusted according to the specific conditions of different reservoirs. For example, in a large reservoir, the following steps can be followed: 1. Obtain data on fish community structure: Using sonar technology and cage sampling, a comprehensive survey is conducted every quarter to obtain information on fish species, quantity, and body length distribution.

[0156] 2. Establish a database: Use the PostgreSQL database system to establish a database containing information on fish population dynamics, age structure, and biomass. The database is updated weekly to record the changing trends of fish community structure.

[0157] 3. Generate strategy: A multi-objective optimization algorithm based on machine learning is used to generate a fish release and fishing strategy once a month. The strategy includes the release quantity, release time and area of ​​each fish, as well as fishing limits and no-fishing areas.

[0158] 4. Perform aquatic vegetation restoration: According to the generated strategy, submerged plants such as black algae and Vallisneria are planted in the shallow water area of ​​the reservoir (water depth of 3-5 meters). Floating leaf plants such as lotus and water lily are planted in the area with a water depth of 1-3 meters.

[0159] 5. Place artificial fish nests: In areas with a water depth of 5-10 meters, 100 artificial fish nests made of environmentally friendly plastic materials are placed per square kilometer. The design of the fish nests imitates the structure of overgrown aquatic plants and is suitable for spawning of a variety of fish.

[0160] Through this systematic approach, the present application can more comprehensively and accurately assess and improve the ecological health of the reservoir. Compared with the prior art, the present application can timely adjust the fish release and fishing strategies to avoid ecosystem imbalance by real-time monitoring of changes in the fish community structure. Not only does it focus on water quality parameters, but it also considers multiple ecological factors such as fish community structure, aquatic vegetation, and artificial fish nests, which can more comprehensively improve the reservoir ecosystem. By establishing a detailed fish community structure database, precise restoration strategies can be formulated for specific fish or specific areas to improve the efficiency of ecological restoration. Through the restoration of aquatic vegetation and the placement of artificial fish nests, a better habitat and breeding environment is provided for fish, enhancing the self-repair and regulation capabilities of the ecosystem.

[0161] In some of the above-mentioned embodiments, during the implementation of the present application, there is still the problem of how to generate scientific and reasonable fish release and fishing strategies based on the reservoir fish community structure database.

[0162] In this regard, the present application further proposes to obtain the fish population quantity, age structure and biomass data in the reservoir fish community structure database; calculate the target fish community structure according to the preset reservoir ecological balance model; compare the target fish community structure with the current fish community structure to obtain difference data; based on the difference data, combined with reservoir water quality parameters and seasonal factors, generate a fish release plan including fish release type, quantity and time; based on the difference data, generate fishing restrictions including fish fishing type, quantity and regional restrictions; integrate the fish release plan and fishing restrictions to form a fish release and fishing strategy.

[0163] This step first obtains the fish population quantity, age structure and biomass data in the reservoir fish community structure database. These data reflect the current fish community status of the reservoir and provide a basis for subsequent analysis and decision-making.

[0164] Next, the target fish community structure is calculated based on the preset reservoir ecological balance model. This model takes into account various factors of the reservoir ecosystem, such as water quality, food chain relationships, habitat conditions, etc., to determine an ideal fish community structure that can maintain ecological balance.

[0165] Then, the target fish community structure is compared with the current fish community structure to obtain the difference data. This step can clearly identify the gap between the current fish community and the ideal state, providing a basis for subsequent adjustments.

[0166] Based on the difference data, combined with reservoir water quality parameters and seasonal factors, a fish stocking plan including fish species, quantity and time is generated. This plan takes into account the impact of water quality and season on fish survival and reproduction, ensuring that the released fish can adapt to the reservoir environment and contribute to the restoration of ecological balance.

[0167] At the same time, based on the difference data, fishing restrictions including fish species, quantity and area restrictions are generated. These restrictions are aimed at controlling the over-breeding of certain fish and maintaining a reasonable population structure.

[0168] Finally, the fish stocking plan and fishing restrictions are integrated into a fish stocking and fishing strategy. This comprehensive strategy balances fish stocking and fishing to achieve the goal of regulating the reservoir ecosystem.

[0169] In practical applications, this method can be implemented in the following ways: First, using underwater sonar systems and regular sampling surveys, we obtain data on the number, length distribution, and biomass of various fish species in the reservoir. These data are stored in a real-time updated database to form a reservoir fish community structure database.

[0170] Next, based on the theory of ecosystem model, a reservoir ecological balance model was established that included water quality parameters, plankton biomass, aquatic vegetation coverage, etc. The model continuously optimized parameters through machine learning algorithms based on historical data and expert knowledge to accurately predict the impact of different fish compositions on the reservoir ecosystem.

[0171] When comparing the current and target fish community structures, a multidimensional analysis method is used. For example, not only the overall biomass is considered, but also the proportion of key species, the balance of each level of the food chain, etc. The difference data can be represented by a multidimensional vector, with each dimension representing the difference of a key indicator.

[0172] When generating a fish stocking plan, the system takes into account seasonal changes in water quality parameters such as water temperature and dissolved oxygen. For example, in the spring when water temperatures rise, more herbivorous fish are stocked to control excessive growth of aquatic plants. At the same time, the system also considers the source and quality of the fry to ensure that the fish stocked have good genetic diversity.

[0173] For fishing restrictions, the system will formulate detailed fishing plans based on the growth cycle and breeding season of fish. For example, fishing bans may be implemented during the breeding season of certain fish, or a rotation system may be implemented in specific areas. At the same time, fishing restrictions will also take into account the economic benefits of fisheries, and try to meet reasonable fishing needs while protecting the ecology.

[0174] The final fish stocking and harvesting strategy is presented in a visual form, including timelines, map annotations, and quantity charts, which are easy for managers to understand and implement. The strategy also includes contingency plans to deal with unexpected situations, such as sudden changes in water quality or outbreaks of fish diseases.

[0175] By implementing this method, reservoir management departments can more accurately control the structure of fish communities, thereby effectively regulating the reservoir ecosystem. The advantage of this method is that it takes into account multiple factors, including ecological balance, seasonal changes, and economic needs, and can formulate more scientific and reasonable fish management strategies. Compared with traditional empirical management methods, this data- and model-based approach can respond to ecological changes more quickly and adjust the structure of fish communities more accurately, thereby more effectively maintaining the ecological health of reservoirs.

[0176] In some of the above-mentioned embodiments, during the implementation of the present application, there are still problems such as difficulty in comprehensively considering multiple ecological stress factors, lack of rapid, comprehensive and accurate assessment capabilities for complex aquatic ecosystems, and difficulty in achieving a balance between ecological security, water supply needs and public concerns.

[0177] In this regard, this application further proposes a method for calculating the reservoir ecological health assessment index. This method uses a comprehensive calculation formula and takes into account multiple influencing factors to more comprehensively and accurately assess the ecological health status of the reservoir.

[0178] The calculation formula for the reservoir ecological health assessment index proposed in this application is: REHM = [Σ(Wi × Pi × Si) × D × T] × E × R The formula takes into account multiple factors, including the weight of each indicator, the weight of monitoring points, seasonal adjustment, depth correction, time response, ecological pressure and risk adjustment, etc. This multi-factor comprehensive assessment method can more comprehensively reflect the health status of the reservoir ecosystem.

[0179] Specifically, the parameters and functions in the formula have the following meanings and functions: Wi represents the weight coefficient of each indicator, ranging from 0 to 1, and is used to adjust the importance of different indicators in the evaluation.

[0180] Pi represents the weight of the monitoring point, with a value range of 0-1, taking into account the representativeness and importance of different monitoring points.

[0181] Si is the seasonal adjustment factor, ranging from 0.8 to 1.2, which is used to reflect the impact of seasonal changes on the reservoir ecosystem.

[0182] D is the depth correction factor, ranging from 0.7 to 1.0, which takes into account the impact of water depth on the ecosystem.

[0183] T is the time response function, expressed as T = 1 - e^(-t1 / τ), where τ is the characteristic time and t1 is the interval between the current time point and the reference time. This function is used to reflect the response time of the ecosystem to environmental changes.

[0184] E is the ecological pressure function, expressed as E = 1 / (1 + e^(-k(MP + BA + IS))), where k is the ecological pressure response coefficient. This function comprehensively considers ecological pressure factors such as microplastic pollution (MP), cyanobacterial biomass (BA) and invasive species impact (IS).

[0185] R is the risk adjustment factor, expressed as R = min(1, max(0.5, 1 - Σ(δiRFi))), where δi is the weight of the ith risk factor and RFi is the risk value of the ith risk factor. This factor is used to adjust the impact of potential risks on ecological health assessment.

[0186] Furthermore, it also includes: Microplastic pollution index MP = Σ(μ1C1 + μ2C2 + ... + μnCn), where μi is the impact weight coefficient of the i-th microplastic and Ci is the concentration of the i-th microplastic. This index reflects the extent and impact of microplastic pollution in reservoirs.

[0187] The cyanobacterial biomass index BA = V × N × (1 + sin(2πt2 / 365)), where V is the volume of cyanobacteria, N is the number density of cyanobacteria, and t2 is the current time point. This index takes into account the seasonal changes in cyanobacterial biomass.

[0188] The impact of invasive species IS = Σ(ρiPi × Ai), where ρi is the ecological impact coefficient of the i-th invasive species and Ai is the distribution area of ​​the i-th invasive species. This index reflects the impact of invasive species on reservoir ecosystems.

[0189] Through this calculation method, this application can more comprehensively and accurately assess the ecological health of the reservoir. This method not only takes into account traditional water quality parameters and biological indicators, but also introduces emerging ecological stress factors such as microplastic pollution, cyanobacteria outbreaks and invasive alien species, thereby improving the comprehensiveness and accuracy of the assessment.

[0190] In practical applications, the weights and value ranges of various parameters can be adjusted according to the characteristics and management needs of specific reservoirs. For example, for a reservoir where cyanobacteria often break out, the weight of the cyanobacterial biomass index BA in the ecological pressure function E can be appropriately increased. Or, for a reservoir located near a city, more attention may need to be paid to microplastic pollution, so the weight of the microplastic pollution index MP can be increased.

[0191] As a preferred implementation, a specific example can be provided to illustrate the application of the calculation method: Suppose in a large reservoir, after a period of monitoring, the following data are obtained: Wi (dissolved oxygen index weight) = 0.3, Pi (main monitoring point weight) = 0.8, Si (summer adjustment factor) = 1.1 D (correction factor for an average water depth of 20 meters) = 0.9 t1 (time interval since last evaluation) = 30 days, τ (characteristic time) = 60 days k (ecological pressure response coefficient) = 0.5 MP (microplastic pollution index) = 0.2, BA (cyanobacteria biomass index) = 0.3, IS (invasive species impact) = 0.1 δ1 (water quality mutation risk weight) = 0.4, RF1 (water quality mutation risk value) = 0.2 Substituting these data into the formula, we get: T = 1 - e^(-30 / 60) ≈ 0.39 E = 1 / (1 + e^(-0.5(0.2 + 0.3 + 0.1))) ≈ 0.57 R = min(1, max(0.5, 1 - 0.4 × 0.2)) = 0.92 The final reservoir ecological health assessment index REHM = [0.3 × 0.8 × 1.1 × 0.9 × 0.39] × 0.57 × 0.92 ≈ 0.0474 The calculated result can be compared with the preset health threshold to determine the ecological health of the reservoir. For example, if 0.05 is set as the passing line, the ecological health of the reservoir is slightly below the passing level and corresponding improvement measures need to be taken.

[0192] Compared with the prior art, the calculation method of this application takes into account a variety of influencing factors, including traditional water quality parameters, biological indicators, and emerging ecological stressors such as microplastic pollution and invasive species. By adjusting the weights and value ranges of each parameter, it can adapt to the characteristics and management needs of different types of reservoirs. The introduction of time response functions and seasonal adjustment factors can reflect the dynamic changes of reservoir ecosystems. Through mathematical models, various influencing factors can be quantified more accurately and the accuracy of evaluation results can be improved. Through the introduction of risk adjustment factors, potential ecological risks can be reflected in a timely manner, providing early warning information for reservoir management.

[0193] Therefore, the reservoir ecological health assessment index calculation method proposed in this application can better solve the problem of rapid, comprehensive and accurate assessment of complex water ecosystems, provide more reliable decision-making basis for reservoir management departments, and help achieve a better balance between ecological security, water supply demand and public concerns.

[0194] Second, refer to Figure 2 , the present application further proposes a reservoir water ecological health diagnosis system, the system comprising: Data acquisition module 210, for acquiring water quality parameters, biological indicators and environmental data collected by distributed sensors in the reservoir; A data fusion module 220 is used to generate reservoir ecological health assessment data through a data fusion algorithm based on water quality parameters, biological indicators and environmental data; An assessment index calculation module 230 is used to calculate the reservoir ecological health assessment index based on the reservoir ecological health assessment data; An assessment result generating module 240 is used to generate a reservoir ecological health status assessment result according to the reservoir ecological health assessment index; A state judgment module 250 is used to judge whether the ecological state of the reservoir is abnormal according to the evaluation result of the ecological health state of the reservoir; The early warning module 260 is used to generate an early warning signal when the ecological status of the reservoir is abnormal.

[0195] By adopting data fusion and comprehensive evaluation index calculation technology, a comprehensive and accurate assessment of the ecological health status of the reservoir is achieved. It has the advantages of comprehensively considering multiple ecological stress factors and quickly and accurately evaluating complex water ecosystems.

[0196] In addition, in some preferred implementations, a reservoir water ecological health diagnosis system proposed in the present application can perform any step of the above method.

[0197] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A reservoir water ecological health diagnosis method, characterized in that: The method includes: Access water quality parameters, biological indicators and environmental data collected by distributed sensors in reservoirs; Generate reservoir ecological health assessment data through data fusion algorithm based on the water quality parameters, biological indicators and environmental data; Calculating the reservoir ecological health assessment index based on the reservoir ecological health assessment data; Generating a reservoir ecological health status assessment result according to the reservoir ecological health assessment index; According to the reservoir ecological health status assessment results, determine whether the reservoir ecological status is abnormal; When the ecological status of the reservoir is abnormal, an early warning signal is generated.

2. A reservoir water ecological health diagnosis method according to claim 1, characterized in that: The step of generating reservoir ecological health assessment data through a data fusion algorithm based on the water quality parameters, biological indicators and environmental data includes: Obtaining the temporal and spatial distribution information of the water quality parameters, biological indicators and environmental data; According to the spatiotemporal distribution information, the water quality parameters, biological indicators and environmental data are subjected to spatiotemporal interpolation processing; Based on the data processed by the spatiotemporal interpolation, reservoir ecological health assessment data is generated through a multi-source data fusion algorithm.

3. A reservoir water ecological health diagnosis method according to claim 2, characterized in that: The step of performing spatiotemporal interpolation processing on the water quality parameters, biological indicators and environmental data according to the spatiotemporal distribution information comprises: The sampling time interval and sampling point spacing for obtaining the water quality parameters, biological indicators and environmental data; Calculating the spatiotemporal interpolation coefficients of the water quality parameters, biological indicators and environmental data according to the sampling time interval and the sampling point spacing; According to the spatiotemporal interpolation coefficients, the water quality parameters, biological indicators and environmental data are subjected to spatiotemporal interpolation calculations to obtain the water quality parameters, biological indicators and environmental data after interpolation processing.

4. A reservoir water ecological health diagnosis method according to claim 3, characterized in that: The step of calculating the spatiotemporal interpolation coefficients of the water quality parameters, biological indicators and environmental data according to the sampling time interval and the sampling point spacing includes: Obtain historical trends of water quality parameters, biological indicators and environmental data; Calculate the change rate of the water quality parameters, biological indicators and environmental data at different sampling time intervals according to the historical change trend; According to the change rate, calculating the spatial correlation of the water quality parameters, biological indicators and environmental data at different sampling point spacings; Generate a temporal interpolation coefficient and a spatial interpolation coefficient according to the change rate and the spatial correlation; The temporal interpolation coefficient and the spatial interpolation coefficient are combined to obtain the temporal and spatial interpolation coefficient.

5. A reservoir water ecological health diagnosis method according to claim 4, characterized in that: The step of calculating the change rate of the water quality parameters, biological indicators and environmental data at different sampling time intervals according to the historical change trend includes: obtaining historical data sets of said water quality parameters, biological indicators and environmental data; Performing time series analysis on the historical data set to identify periodic change patterns and long-term change trends; Establishing a time series prediction model based on the periodic change pattern and long-term change trend; Using the time series prediction model, predicting data values ​​at different sampling time intervals; The difference between adjacent predicted data values ​​is calculated and divided by the corresponding time interval to obtain the change rate of the water quality parameters, biological indicators and environmental data at different sampling time intervals.

6. A reservoir water ecological health diagnosis method according to claim 1, characterized in that: The step of calculating the reservoir ecological health assessment index according to the reservoir ecological health assessment data comprises: Obtaining water quality parameters, biological indicators and remote sensing data from the reservoir ecological health assessment data; Calculating a water quality parameter index according to the water quality parameters, wherein the water quality parameter index includes a weighted sum of dissolved oxygen, pH value, total nitrogen, total phosphorus and temperature; Calculating a bio-indicator score based on the bio-indicator, wherein the bio-indicator score comprises a weighted sum of algae, fish, and plankton; Calculating a remote sensing data index based on the remote sensing data, wherein the remote sensing data index comprises a weighted sum of a normalized water index, chlorophyll a, and turbidity; The reservoir ecological health assessment index is calculated based on the water quality parameter index, biological indicator score and remote sensing data index.

7. A reservoir water ecological health diagnosis method according to claim 1, characterized in that: The method further comprises: Obtaining reservoir fish community structure data based on the reservoir ecological health status assessment results; Establishing a reservoir fish community structure database according to the reservoir fish community structure data; generating fish stocking and fishing strategies based on the reservoir fish community structure database; Perform aquatic vegetation restoration in accordance with described fish stocking and harvesting strategies; According to the execution results of the aquatic vegetation restoration, artificial fish nests are placed.

8. A reservoir water ecological health diagnosis method according to claim 7, characterized in that: The step of generating fish release and fishing strategies according to the reservoir fish community structure database comprises: Obtaining fish population quantity, age structure and biomass data from the reservoir fish community structure database; According to the preset reservoir ecological balance model, the target fish community structure is calculated; Comparing the target fish community structure with the current fish community structure to obtain difference data; Based on the difference data, combined with reservoir water quality parameters and seasonal factors, a fish stocking plan including fish stocking type, quantity and time is generated; generating fishing restrictions including fish catch type, quantity and area restrictions based on the difference data; The fish stocking plan and the fishing restrictions are integrated to form the fish stocking and fishing strategy.

9. A reservoir water ecological health diagnosis method according to claim 1, characterized in that: The calculation formula for calculating the reservoir ecological health assessment index based on the reservoir ecological health assessment data is: REHM = [Σ(Wi × Pi × Si) × D × T] × E × R in, DREHM is the reservoir ecological health assessment index; Wi is the weight coefficient of each indicator, and its value range is 0-1; Pi is the weight of the monitoring point, ranging from 0 to 1; Si is the seasonal adjustment factor, ranging from 0.8 to 1.2; D is the depth correction coefficient, ranging from 0.7 to 1.0; T is the time response function, T = 1 - e^(-t1 / τ); τ is the characteristic time; t1 is the time interval between the current time point and a certain reference time; E is the ecological pressure function, E = 1 / (1 + e^(-k(MP + BA + IS))); k is the ecological pressure response coefficient; R is the risk adjustment factor, R = min(1, max(0.5, 1 - Σ(δiRFi))); MP is the microplastic pollution index, MP = Σ(μ1C1 + μ2C2 + ... + μnCn); μi is the impact weight coefficient of the i-th type of microplastics, i=1...n; Ci is the concentration of the i-th type of microplastics, i=1...n; BA is the cyanobacterial biomass index, BA = V × N × (1 + sin(2πt2 / 365)); V is the volume of cyanobacteria; N is the number density of cyanobacteria; t2 is the current time point; IS is the impact of invasive species, IS = Σ(ρiPi × Ai); ρi is the ecological impact coefficient of the i-th invasive species; δi is the weight of the ith risk factor; Ai is the distribution area of ​​the i-th invasive species; RFi is the risk value of the ith risk factor.

10. A reservoir water ecological health diagnosis system, characterized in that: The system includes: Data acquisition module, used to obtain water quality parameters, biological indicators and environmental data collected by distributed sensors in the reservoir; A data fusion module, used to generate reservoir ecological health assessment data through a data fusion algorithm based on the water quality parameters, biological indicators and environmental data; An assessment index calculation module, used to calculate the reservoir ecological health assessment index based on the reservoir ecological health assessment data; An assessment result generating module, used to generate a reservoir ecological health status assessment result according to the reservoir ecological health assessment index; A state judgment module, used to judge whether the ecological state of the reservoir is abnormal according to the ecological health state assessment result of the reservoir; The early warning module is used to generate early warning signals when the ecological status of the reservoir is abnormal.

Images