WO2018214190A1 - 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法 - Google Patents

一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法 Download PDF

Info

Publication number
WO2018214190A1
WO2018214190A1 PCT/CN2017/089012 CN2017089012W WO2018214190A1 WO 2018214190 A1 WO2018214190 A1 WO 2018214190A1 CN 2017089012 W CN2017089012 W CN 2017089012W WO 2018214190 A1 WO2018214190 A1 WO 2018214190A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
monitoring
lake
water
time
Prior art date
Application number
PCT/CN2017/089012
Other languages
English (en)
French (fr)
Inventor
秦伯强
吴挺峰
朱广伟
张运林
李未
Original Assignee
中国科学院南京地理与湖泊研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国科学院南京地理与湖泊研究所 filed Critical 中国科学院南京地理与湖泊研究所
Priority to US16/329,943 priority Critical patent/US11402362B2/en
Priority to PCT/CN2017/089012 priority patent/WO2018214190A1/zh
Publication of WO2018214190A1 publication Critical patent/WO2018214190A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18Water
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01DMEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00Measuring or testing not otherwise provided for
    • G01D21/02Measuring two or more variables by means not covered by a single other subclass
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/0098Plants or trees

Definitions

  • the invention belongs to the field of environmental monitoring evaluation and data mining technology, in particular to a stereoscopic monitoring and data mining system and method for lake blue algae disaster.
  • a common phenomenon caused by eutrophication of lakes is that under suitable hydrometeorological conditions, many phytoplankton, especially those with buoyancy or exercise ability, will proliferate abnormally and aggregate to form surface blooms, which will trigger a series of Serious ecological problems include: reduced transparency of water, reduced dissolved oxygen, death of aquatic animals and plants, odors emitted by water bodies, decreased biodiversity, and damage to human health through the food chain.
  • the continued global expansion of ecological disasters caused by eutrophication and abnormal growth of cyanobacteria has pose a huge threat to the ecological health and sustainable development of many famous lake ecosystems, including: Lake Erie, USA, Canada Lake Nibe, the Baltic Sea in Europe, Lake Victoria in Africa and Lake Biwa in Japan.
  • Forecasting, and publishing information through the Internet but from its design framework and text description, the main problems of this technology are: (1) does not include satellite receiving antenna, the data obtained is not real-time satellite remote sensing data, so the platform The function of automatic real-time acquisition of satellite remote sensing data cannot be realized; (2) The automatic monitoring station only describes the setting of the chlorophyll sensor in detail, and does not mention the monitoring indicators, arrangement methods and parameter settings of other instruments and equipment, and does not introduce real-time monitoring.
  • the technology only mentions the cyanobacteria biomass model, and does not have the simulation function of the ecological process closely related to the algae life process, such as nutrient salt circulation, sediment erosion suspension and dissolved oxygen dynamics; (4) The technology also has the disadvantage of not being able to fully exploit the value of the data generated by the monitoring. Both remote sensing and automatic monitoring can generate massive amounts of data on cyanobacterial disasters. However, the prior art cannot fully utilize this part of the data for lake research and management services, and it cannot provide valuable and readable cyanobacterial disaster information for the general public.
  • the present invention aims to overcome the deficiencies of the prior art and to provide a stereoscopic monitoring and data mining system and method for cyanobacteria disasters.
  • the real-time cyanobacteria disaster data is collected by integrating satellite remote sensing, automatic monitoring and manual patrol technology, and the database, data assimilation and numerical model are utilized for the problem of large spatial and temporal variation of cyanobacteria blooms.
  • the present invention adopts the following technical solutions:
  • a stereoscopic monitoring and data mining system for lake blue algae disasters comprising a monitoring subsystem and a data processing subsystem;
  • the monitoring subsystem is configured to collect data to be monitored; the remote sensing monitoring subsystem that uses remote sensing for monitoring, the automatic monitoring subsystem that uses automatic monitoring stations for monitoring, and the manual patrol subsystem that manually collects data to be monitored;
  • the remote sensing monitoring subsystem comprises a satellite data receiving antenna, a computer and a power supply system, wherein the power supply system is used for powering the power consumption device, the satellite data receiving antenna receives the satellite data, and transmits the data to the computer, and the remote sensing inversion processing is performed by the computer. Transmitting the processed data to the data processing subsystem via the Internet;
  • the automatic monitoring subsystem is a monitoring network formed by a plurality of automatic monitoring stations connected by a wireless network.
  • the single automatic monitoring station is composed of a water surface supporting system, a power supply system, a safety warning system and a data collecting system; the water surface supporting system is automatically monitored. a load-bearing structure of the station hardware device; the power supply system is configured to supply power to the power consumption device of the automatic monitoring station; the safety warning system is used for safety warning to prevent the automatic monitoring station from being accidentally damaged; and the data acquisition system is used for collecting The data to be monitored obtained from meteorological instruments, hydrological instruments, water quality instruments and video instruments, and the collected data is transmitted to the data processing subsystem through the wireless network;
  • the surface support system is the load-bearing structure of the entire system and can be a buoy body, a trestle bridge or a water platform.
  • the upper part of the support system is exposed to the water surface, and the lower part is fixed at the bottom of the lake to function as a support for the entire system structure.
  • the power supply system is fixed to the surface support system to provide power to other power consuming equipment through cables.
  • the safety alert system is mounted on the surface support system, including beacon lights, fluorescent bands and warning signs.
  • the manual patrol subsystem manually collects data and transmits the data to the data processing subsystem via the Internet;
  • the data processing subsystem is configured to receive and process data acquired by the monitoring subsystem; including a server, an array machine, a computer, a computing workstation, a hardware firewall, a router, a network cable, and a power supply device; the server adopts a dual hot standby mode, and two servers
  • the array machine for data storage implements a heartbeat connection through a three-prong cable. All computers, servers, and computing workstations are connected to the hardware firewall through a network cable, and the hardware firewall is connected to the external network through a connection router;
  • the server adopts a dual-system hot standby mode, and the central server is installed as two servers that are mutually backed up, and only one server runs at a time. When one of the servers running in it fails to start, the other backup server will automatically start and run quickly.
  • the computing workstation is mainly used for the calculation of three-dimensional numerical models.
  • the hardware firewall is mainly used to protect the entire system from attacks, and the router provides network services for the entire system. These devices will be placed in the cabinet and the cabinet power supply will be connected to the uninterruptible power supply. Under normal circumstances, all devices are powered by an external power source. When the external power supply is interrupted unexpectedly, the power supply of the entire system is provided by the uninterruptible power supply.
  • the data processing subsystem After receiving the data, the data processing subsystem performs data backup and data processing on the received data to obtain predictive indicators and risk assessments of lake blue-green algae disasters, and publish them through public platforms.
  • the invention also provides a stereoscopic monitoring and data mining method for cyanobacterial disasters, which comprises the following steps:
  • monitoring indicator data related to the cyanobacteria disaster of the lake to be monitored through remote sensing monitoring, automatic monitoring and manual patrol, including meteorological indicators, hydrological indicators, water quality indicators and video images, and the acquired data is transmitted to the data via the Internet. center;
  • the remote sensing monitoring refers to real-time monitoring by satellite remote sensing; the remote sensing monitoring data acquisition is received by the satellite receiving antenna and transmitted to the remote sensing processing computer, and the satellite data inversion software generates various remote sensing indicators. Thereafter, the data is passed.
  • the computer sharing function in the local area network is implemented to be sent to the server data center;
  • the automatic monitoring refers to that a plurality of automatic monitoring stations are connected through a wireless network to form a monitoring network, and the monitoring indicators are monitored; the automatic monitoring collects meteorological, hydrological and water quality indicators and videos through various types of sensors, and is used by the communication module.
  • the communication network is sent to the server data center, and the data is received by the commercial software installed on the server side;
  • the manual patrol refers to manual detection of indicators; manual patrol refers to pre-set through a predetermined time
  • the fixed monitoring site carries out on-site data and water sample collection, and sends the water sample to the laboratory to detect relevant indicators.
  • the manual inspection data is manually uploaded to the server data center through the website;
  • Remote sensing monitoring, automatic monitoring and manual patrol data transmitted to the server are stored in the server data center in a specific file naming manner to facilitate identification of the original data by various data processing programs;
  • the method for determining the remote sensing monitoring index system is based on the influence of water quality indicators such as water temperature and water color on the emission and receiving spectra of the optical sensors carried by the satellite.
  • the selection is closely related to the cyanobacterial disaster, and can also be used from the spectral signals by using the inversion algorithm.
  • Water quality indicators for extracting effective information are used as satellite inversion indicators, including: water temperature, transparency, suspended matter (SS), chlorophyll a, cyanobacterial bloom area and intensity.
  • the method for determining the automatic monitoring index system is based on the research on the maturity of meteorological, hydrological and water quality sensor technologies on the market, and selecting indicators supported by sensor technology that are closely related to cyanobacterial disasters and have higher maturity as automatic monitoring indicators.
  • meteorological indicators wind speed, wind direction, air pressure, temperature, humidity, solar radiation and rainfall
  • hydrological indicators three-dimensional flow profile, water depth, effective wave height and cycle, etc.
  • water quality indicators water temperature, dissolved oxygen, turbidity, Conductivity, redox potential, phycocyanin, chlorophyll, etc.
  • video image
  • the applicable manual patrol indicator system is determined as follows: indicators that cannot be used for remote sensing inversion and automatic monitoring, and indicators closely related to cyanobacteria disasters will be obtained through manual patrol, including: total nitrogen, dissolved total nitrogen, Ammonia nitrogen, nitrate nitrogen, nitrite nitrogen, total phosphorus, dissolved total phosphorus, orthophosphate and other cyanobacterial-derived pollutants.
  • Remote sensing monitoring, automatic monitoring and manual inspection data are stored in the server in a specific file naming manner to facilitate identification of the original data by various data processing programs.
  • the data center performs data backup and data preprocessing on the received data, and the data preprocessing includes the following steps:
  • the data is checked according to the set data collection time interval. If the data is interrupted, the data is interpolated by the time interpolation method;
  • the data of the cloud coverage area is filled by spatial interpolation
  • the data After the data is preprocessed, it is transmitted to the database storage together with the original data received by the data center;
  • the material migration transformation model is superimposed on the hydrodynamic model of the lake to be monitored, and the equations of the two models are coupled and calculated; wherein the scalar calculated by the material migration conversion model includes illumination, suspended matter, algae growth, nutrient salt circulation and dissolved oxygen. ;
  • the three-dimensional numerical model used may be an existing model such as FVCOM, ELCOM-CAEDYM, and EFDC, or a model independently developed based on environmental fluid mechanics theory;
  • step (2) the data stored in the database is classified and stored, as follows:
  • the data generated by the three-dimensional numerical model is stored in the data table with time as a node;
  • the image or video is stored in an array machine, and a data table is created in the database to record the path of the image or video.
  • the fields include number, time, and image/video path to access the image or video in an indexed manner.
  • the remote sensing monitoring, automatic monitoring and manual inspection data are stored in the server in a specific file naming manner, and the original data is recognized by various data processing programs.
  • the time interpolation method adopts linear interpolation, spline function interpolation or piecewise interpolation;
  • the spatial interpolation method adopts the neighboring point method, the Kriging method or the inverse distance weighting method
  • the basis for the determination and processing of outliers is trend checking, expert experience or numerical comparison.
  • the time interpolation method adopts linear interpolation, and the algorithm is as follows:
  • the spatial interpolation method uses an inverse distance weighting method, and the algorithm is as follows:
  • the outlier determination is performed by the 5-fold variance method.
  • the specific method is as follows: the m-th measured data a m and the five data before and after it are averaged and varianced:
  • the three-dimensional numerical model used in the step (3) may be an existing model such as FVCOM, ELCOM-CAEDYM, and EFDC, or may be a model independently developed based on the theory of environmental fluid mechanics; the present invention preferably provides A self-developed model, as follows:
  • v 0 , m 0 , m 1 values are 5.0 ⁇ 10 -6 m 2 /s, 0.1 and -1 respectively;
  • l is the Prandtl length;
  • R i is the Richardson number, reflecting the fluid stability condition, and its expression The formula is:
  • ⁇ a and ⁇ s represent air density and surface water density, respectively;
  • C WD is wind drag coefficient;
  • WS is wind speed at 10 m above water meter;
  • C SD is lake bottom friction coefficient;
  • ⁇ PAR (I, J, K), ⁇ 0, PAR represent the total attenuation coefficient and the pure water attenuation coefficient respectively;
  • ⁇ 1, PAR , ⁇ 2, PAR represent the algae ratio attenuation coefficient and the non-algae particle specific attenuation coefficient, respectively ;
  • CHLA(I,J,K) is the phytoplankton biomass expressed as chlorophyll a concentration;
  • u,v,w is the three-dimensional flow velocity of the new time step;
  • S suspended Mobility concentration;
  • w s is the sedimentation rate of suspended solids;
  • J 0 is the flux term of water and soil interface, including erosion flux and sedimentation flux;
  • C i is the concentration of the i-th substance;
  • the finite difference method is used to discretize the above equations in a rectangular grid: spatially discrete using upwind style; horizontal and temporal difference schemes are explicit, vertical difference scheme is implicit; chasing method is used to solve hyperlarge sparse matrices ;
  • is the total growth rate
  • KM is the non-pastoral mortality
  • KS is the floating rate
  • ZP is the zooplankton filter rate
  • CHLA i, j, k is the phytoplankton biomass of the i, j, k grid
  • ⁇ a , ⁇ f , ⁇ p and the following formulas ⁇ ZP , ⁇ KB , ⁇ s , ⁇ d , ⁇ so are temperature influence factors, and the expressions are ⁇ a T-20 , ⁇ f T -20 , ⁇ p T-20 , ⁇ ZP T-20 , ⁇ KB T-20 , ⁇ s T-20 , ⁇ d T-20 and ⁇ so T-20 , where ⁇ a , ⁇ p , ⁇ ZP , ⁇ KB , ⁇ s , ⁇ d and ⁇ so are temperature multipliers;
  • ⁇ zp is the growth rate of zooplankton
  • KCHLA is the semi-saturated parameter of zooplankton
  • BFISH and FISH respectively represent fish drainage rate and fish biomass
  • ZOOP represents zooplankton biomass
  • FMRP, FMRN, FMDP and FMDN are the phosphorus conversion rate, metabolite nitrogen conversion rate, death residue phosphorus conversion rate and death residue nitrogen conversion rate of algae metabolites respectively; KD and KM are algae metabolic rate and mortality, respectively; RPJ The static release rate of phosphorus and nitrogen in sediments and RNJ respectively; the dynamic release rate of phosphorus and nitrogen in sediments of RPD and RND; the flux of suspended matter in SEDF soil and water interface; the absorption rate of phosphorus and nitrogen by algae in ZDP and ZDN respectively; the DPS in KPS and KNS And the sedimentation rate of DTN;
  • KOD atmospheric reoxygenation rate
  • DOSAT saturated DO
  • H s characteristic wave height
  • ZHY algae respiratory oxygen consumption
  • RO sediment oxygen consumption rate
  • ZHD and ZHR algae death and metabolites BOD production rate
  • KB is the rate of BOD degradation.
  • the three-dimensional numerical model reads the initialization data and the boundary condition data at a certain point in time from the database to improve the calculation speed and efficiency, and can adopt the Monte Carlo method according to the measured data in the previous simulation period. Optimize the model parameter combination to improve the prediction accuracy of the current simulation model; after completing the required data input, the model automatically activates the current round simulation and returns the current calculation result to the database; after finishing the above process The model goes to sleep and waits for the next round of simulation. This can achieve autonomous rolling forecasts.
  • the lake blue cyanobacteria disaster prediction and warning information is displayed on a public platform
  • the public platform may use an internet-based software platform to use the relevant website for information release
  • the cyanobacteria disaster information release website is an internet-based software.
  • the platform is a dialogue window for managing stereoscopic monitoring and data mining systems, and realizes information interaction between users and systems through a website form.
  • remote sensing monitoring can collect surface-distributed indicators
  • automatic monitoring can collect time-continuous indicators at very high frequencies
  • manual monitoring can make up for indicators that cannot be collected by the above two technologies.
  • Figure 1 is a schematic flow chart of the method of the present invention
  • FIG. 2 is a distribution diagram of an automatic monitoring station according to Embodiment 1 of the present invention.
  • FIG. 3 is a schematic diagram of a numerical calculation grid division of a model according to Embodiment 1 of the present invention.
  • monitoring indicator data related to the cyanobacteria disaster of the lake to be monitored through remote sensing monitoring, automatic monitoring and manual patrol, including meteorological indicators, hydrological indicators, water quality indicators and video images, and the acquired data is transmitted to the data via the Internet. center;
  • the remote sensing monitoring refers to real-time monitoring by satellite remote sensing
  • the automatic monitoring refers to that a plurality of automatic monitoring stations are connected through a wireless network to form a monitoring network, and the monitoring indicators are monitored in real time;
  • the manual patrol refers to manual detection of indicators
  • Taihu Lake Located in the Yangtze River Delta of China, Taihu Lake is the third largest freshwater lake in China and is currently facing serious eutrophication and cyanobacterial blooms.
  • the water crisis that occurred in Wuxi in 2007 caused huge losses to the people's production and life in the Taihu Lake.
  • step (1) carry out on-site and laboratory comparison tests on all indicators closely related to cyanobacteria blooms, and fully investigate the development of relevant sensor technologies, and give remote sensing monitoring, automatic monitoring and manual inspection for each indicator.
  • remote sensing monitoring indicators include: effective wave height, water temperature, turbidity Degree, transparency, chlorophyll a, water bloom area, water bloom intensity
  • automatic monitoring indicators include: wind speed, wind direction, air pressure, temperature, humidity, solar radiation and rainfall, three-dimensional flow profile, water depth, effective wave height and cycle, water temperature, dissolution Oxygen, turbidity, conductivity, redox potential, phycocyanin, chlorophyll, video images
  • artificial patrol indicators include: various nitrogen and phosphorus concentrations, chlorophyll a, phytoplankton, toxic and harmful substances, algal toxins and benthic animals.
  • the preprocessing includes the following steps:
  • spatial interpolation is used to fill the missing data of the cloud coverage area.
  • the spatial interpolation method can adopt the neighboring point method.
  • the Kriging method or the inverse distance weighting method; in this embodiment, the inverse distance weighting method is used to implement spatial interpolation. Assuming that the index value of the spatial point coordinates (x 0 , y 0 ) is missing C (x 0 , y 0 ), set a search radius so that there are at least 3 data points in this radius; then use these known data Point data to find the indicator value of the unknown point:
  • C(x 1 , y 1 ), C(x 2 , y 2 ), ..., C(x n , y n ) represent the measured index values at the coordinate points in parentheses; d 1 , d 2 ,..., d n represents the linear distance from the coordinate point in the parenthesis to the spatial point coordinate (x 0 , y 0 ), n ⁇ 3.
  • the basis of the abnormal value determination and processing is trend test, expert experience or numerical comparison.
  • the previous data of the abnormal data is used instead of the abnormal data; in this embodiment, the data is abnormally determined by using the 5x variance method.
  • the data generated by the three-dimensional numerical model is stored in the data table with time as a node;
  • the image or video is stored in an array machine, and a data table is created in the database to record the path of the image or video.
  • the fields include number, time, and image/video path to access the image or video in an indexed manner.
  • the data table name is the monitoring station name; the data table field is the indicator name; the data record is the measured data value stored in chronological order.
  • the storage process of the water temperature, chlorophyll a and water depth of the automatic monitoring station EMB16 as shown in FIG. 2 is: 1) first establish a data table named after EMB16; 2) the field names of the table are: number, time, water temperature , chlorophyll a and water depth, etc.; 3) The first row of the table stores the first data recorded by EMB16.
  • the satellite image only stores the path of the image storage in the data table; the two-dimensional data obtained by the satellite image inversion is directly stored in the data table.
  • the satellite image is stored in the folder named "satellite picture" in the array machine.
  • the data table is created and named as "satellite picture path"; the fields include: number, time, picture path, remark; data
  • Each record in the table corresponds to a satellite image acquired at a certain time. After satellite inversion, usually each pixel will correspond to one data.
  • a satellite image with a north-south span of 250*M meters and an east-west span of 250*N meters with a resolution of 250m will generate an array of M rows and N columns after inversion of its water temperature index.
  • the array is stored as: create a data table and name it "inversion water temperature"; the fields include: number, water layer, time, water temperature 1, water temperature 2, ..., water temperature N; the first row of the table Stores the first, second, ..., N columns of data in the first row of the array, the second row of the table stores the first, second, ..., N columns of data in the second row of the array, and so on, until the entire The storage of the two-dimensional array; after completing the water temperature data inversion by the satellite image at the first moment, the satellite inversion value of the second time point is stored, and so on, wherein the water layer and time can mark different moments Point satellite inversion data.
  • the data generated by the 3D numerical model this will be a 3D array.
  • the horizontal direction of the study lake is divided into M rows and N columns, and the water depth direction is divided into K layers, and for the simulated water temperature, it is TEMP (M, N, K).
  • the storage of the three-dimensional array is to realize the storage of K two-dimensional arrays at one time: the data table is named “simulated water temperature”; the fields include: number, water layer, time, simulated water temperature 1, simulating water temperature 2, ..., simulating water temperature N; each water layer is stored in the same manner as the inversion water temperature; after completing the first water layer two-dimensional array storage, the second water value of the second water layer is stored , and so on, until the K-layer two-dimensional value storage is completed. After the 3D array storage of the first time point is completed, the 3D value storage of the second time point is performed, and so on, wherein the water layer and the time can mark the model simulation data of the same water layer and different time points.
  • the material migration transformation model is superimposed on the hydrodynamic model of the lake to be monitored, and the equations of the two models are coupled and calculated; wherein the scalar calculated by the material migration conversion model includes illumination, suspended matter, algae growth, nutrient salt circulation and dissolved oxygen. ;
  • the self-constructed model is used to describe the equations of the lake water movement as follows:
  • T temperature
  • S h the heat entering the system externally
  • C p the heat capacity of the water body
  • the water body density
  • ⁇ 0 the reference density of the water body
  • a h is the horizontal eddy viscosity coefficient, and Taihu Lake takes 5m 2 /s
  • a v is the vertical eddy viscosity coefficient, which is defined by the following formula:
  • v 0 , m 0 , m 1 values are 5.0 ⁇ 10 -6 m 2 /s, 0.1 and -1 respectively;
  • l is the Prandtl length;
  • R i is the Richardson number, reflecting the fluid stability condition, Its expression is:
  • C WD is the wind drag coefficient, here is 0.001;
  • WS is the wind speed at 10m above the water meter ( m/s);
  • C SD is the coefficient of friction at the bottom of the lake, taking 0.003.
  • the internal and external modes are generated; then the finite difference is used to discretize the internal and external modes, wherein the horizontal and temporal difference formats are explicit and vertical.
  • the difference format is implicit, and the low-pass filter is used to smooth the surface displacement in the time direction.
  • the chasing method is used to solve the super-large sparse matrix.
  • the calculation grid is set to divide the calculation domain into 69 ⁇ 69 grids in a horizontal direction using a rectangular grid with a side length of 1000 m; and divide into 5 layers in the vertical direction.
  • the time step is 30s.
  • ⁇ PAR (I, J, K), ⁇ 0, PAR represent the total attenuation coefficient and the pure water attenuation coefficient respectively; ⁇ 1, PAR , ⁇ 2, PAR represent the algae ratio attenuation coefficient and the non-algae particle specific attenuation coefficient, respectively ;CHLA (I, J, K) phytoplankton biomass expressed as chlorophyll a concentration; SED (I, J, K) non-algal particulate matter concentration.
  • ZOOP chlorophyll a, zooplankton
  • DTP available phosphorus
  • DTN available nitrogen
  • DO dissolved oxygen
  • BOD biochemical oxygen demand
  • the finite difference method is used to discretize the above equations in a rectangular grid: spatially discrete using upwind style; horizontal and temporal difference schemes are explicit, vertical difference scheme is implicit; chasing method is used to solve hyperlarge sparse matrices .
  • the biochemical term involved in equation (17) is calculated by the following method.
  • FMRP, FMRN, FMDP and FMDN are the phosphorus conversion rate, metabolite nitrogen conversion rate, death residue phosphorus conversion rate and death residue nitrogen conversion rate of algae metabolites respectively;
  • KD and KM are algae metabolic rate and mortality, respectively;
  • RPJ The static release rates of phosphorus and nitrogen in sediments and RNJ respectively;
  • RPD and RND are the dynamic release rates of phosphorus and nitrogen in sediments respectively;
  • SEDF is the flux of suspended matter in water-soil interface;
  • ZDP and ZDN respectively indicate the absorption rate of phosphorus and nitrogen by algae;
  • KPS and KNS is the sedimentation rate of DTP and DTN.
  • the three-dimensional numerical model reads the initialization data and boundary condition data of a certain point in time from the database, and optimizes the model parameter combination by Monte Carlo method according to the measured data in the previous simulation period; after completing the required data input, the model Automatically activate the current round of simulation operations, and return the results of this round of calculations to the database; after the end of the above process, the model is transferred Sleep state, waiting for the next round of simulation operations.
  • the three-dimensional numerical model calculation for the cyanobacteria bloom of Taihu Lake adopts a rectangular grid.
  • the whole lake is divided into 4900 computing grids including land and water in the horizontal plane, as shown in Figure 3.
  • the model reads the measured values of lake wind, lake, water temperature, light, chlorophyll a, nutrient salt, dissolved oxygen and organic matter from the Oracle database at 12:00 on the forecast day, and uses the inverse distance weight interpolation method.
  • 4900 computational grids the value in the terrestrial grid is set to -9999 as the initial concentration field for this round of forecasting.
  • the model parameters of the current forecast are optimized, and the Monte Carlo method is used to optimize the model parameters.
  • the initialization and parameter optimization are completed, it is also necessary to obtain the weather condition of the current forecast period of 3 days.
  • the wind speed, wind direction, temperature, radiation, precipitation, air pressure and relative humidity required for the model for the next 3 days are calculated by the Weather Observation and Prediction Model (WRF).
  • WRF Weather Observation and Prediction Model
  • the Taihu Lake cyanobacteria bloom risk assessment algorithm is based on the spatial distribution of lake wind, lake, water temperature, light, chlorophyll a, nutrient salt, dissolved oxygen and organic matter in the future 3D modeled by three-dimensional numerical model. It is determined that the risk level of cyanobacterial blooms occurs in different places at different times in Taihu Lake, and the risk level is characterized by different colors.
  • the specific implementation is as follows:
  • the public platform can use the Internet-based software platform to publish information through relevant websites.
  • the cyanobacteria disaster information publishing website is an Internet-based software platform that manages stereoscopic monitoring and data.
  • the dialogue window of the mining system is used to realize the information interaction between the user and the system through the website form.
  • the cyanobacteria disaster information publishing website constructed in this embodiment is an internet-based software platform, which is a dialogue window for managing a stereoscopic monitoring and data mining system, and realizes information interaction between the user and the system through a website form.
  • the main functions are: home page, telemetry data Patrol Test data, satellite data, forecast and warning, user center and other functions.
  • the stereoscopic display is implemented in three web pages of telemetry data, patrol data and satellite data, and their respective pages function similarly.
  • the three-dimensional monitoring webpage interacts with the database to display the real-time environment information of the lake collected by the Taihu stereo monitoring system to the public.
  • the main loading controls and reference objects include: database engine, buttons, charts, check boxes, drop-down combo boxes, text boxes, time And WebGIS and so on.
  • the web page can display real-time meteorological, hydrological and water quality monitoring data of a single site on an electronic map, or display the surface monitoring data through a secondary development of the electronic map.
  • the webpage can also provide historical data retrieval, downloading and simple statistical analysis services for users with different rights.
  • the forecasting and warning webpage interacts with the database to display the prediction and warning information of the Taihu Lake cyanobacteria disaster.
  • the main loading controls and reference objects include: database engine, time, button, iWebOffice, text box, drop-down list box, check box, Frame, WebGIS. And Flash and more.
  • the webpage has: 1) information such as a calculation grid, a time step, a forecast period, an initial value, and a boundary condition involved in displaying a three-dimensional numerical model in the form of a web page table, and also lists all model parameter names, functions, and value ranges. And information such as the current value.
  • the user can modify the above model settings through the built-in iWebOffice plug-in; 2) draw the contour map, the web page can retrieve the model simulation data, draw the contour map through WebGIS, and play these in the chronological order of the Flash control.
  • the value-line diagram will form an animation of the temporal and spatial changes of the simulated environmental indicators of the Taihu Lake in the future 3D; 3)
  • the webpage can automatically produce the Taihu Lake water pollution and cyanobacteria monitoring and warning semi-weekly report according to the pre-specified format.
  • the semi-weekly report can also be downloaded via a hyperlink or automatically sent to the specified email address at a specified time.
  • the user center webpage mainly implements functions such as managing users at different levels and publishing news by users themselves.
  • the main loading controls and reference objects include: database engine, buttons, Flash, images, and tables.
  • the web page can set different permissions for all users, including system administrators, advanced users, intermediate users and novice users.
  • the system administrator has all the permissions on the website and can make all the rules.
  • Advanced users can browse all pages of the website, access the database, download monitoring data in batches, publish news and manipulate 3D numerical models.
  • Intermediate users can browse all pages of the website, access the database, and download monitoring data in batches.
  • the novice user is a general visitor and can only view the cyanobacterial disaster information displayed by the website to all the general public.
  • This embodiment takes the Taihu Lake as an example to further describe the system of the present invention.
  • the stereoscopic monitoring and data mining system for the cyanobacteria disaster of the lake comprises a monitoring subsystem and a data processing subsystem, wherein the monitoring subsystem is used for collecting data to be monitored; including a remote sensing monitoring subsystem using remote sensing for monitoring, and utilizing automatic An automatic monitoring subsystem for monitoring by the monitoring station and a manual patrol subsystem for manually collecting data to be monitored;
  • the data processing subsystem is configured to receive and process data acquired by the monitoring subsystem; including a server, an array machine, a computer, a computing workstation, a hardware firewall, a router, a network cable, and a power supply device; the server adopts a dual hot standby mode, and the two services
  • the server and the array machine for data storage implement a heartbeat connection through a three-prong cable. All computers, servers, and computing workstations are connected to the hardware firewall through a network cable, and the hardware firewall is connected to the external network through a connection router;
  • the data processing subsystem After receiving the data, the data processing subsystem performs data backup and data processing on the received data, including data preprocessing, three-dimensional numerical simulation and cyanobacteria disaster assessment, and obtains predictive indicators and risk assessments of lake cyanobacteria disasters, and publishes them through public platforms.
  • the data pre-processing, three-dimensional numerical simulation and cyanobacteria disaster assessment algorithm can refer to the algorithm selected in Embodiment 1.
  • the power supply device is powered by an uninterruptible power supply of the UPS system, and the two servers are located in a local area network under a hardware firewall and a router, and a computer for receiving and processing remote sensing data is also used in the same local area network, and is used for a three-dimensional numerical model.
  • the running computing workstation, the computing workstation and the two servers are in the same local area network, can meet the automatic and fast reading of the database data by the three-dimensional numerical model, and can return the calculation result to the Oracle database, all the devices pass the cable and the UPS system.
  • the UPS system is connected to the civilian AC, and the UPS system provides a stable power supply for the data center.
  • Satellite data on a computer in a local area network is transmitted to the server by writing a remote sensing data receiving program, and stored in the array machine according to a prescribed file naming pattern.
  • the communication module of the automatic monitoring station includes a GRPS module and a CR1000 data collector. As long as the lognet software matching the CR1000 is installed on the server side, the automatic monitoring data can be transmitted to the server side and stored in the array machine according to the specified file naming pattern.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Food Science & Technology (AREA)
  • Medicinal Chemistry (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Wood Science & Technology (AREA)
  • Botany (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法,通过遥感监测、自动监测和人工巡测三种途径获取与湖泊蓝藻灾害相关的监测指标数据,将获取的数据通过互联网传输至数据中心;数据中心对接收的数据进行数据备份和数据预处理,包括时间插值、空间插值和异常判定处理;数据经预处理后传输至数据库存储;根据数据库中存储的数据源进行湖泊的三维数值模型计算,根据数值模型模拟数据,进行待监测湖泊蓝藻灾害风险评估,并在公共平台上展示湖泊蓝藻灾害预测预警信息。该方法和系统实现了对蓝藻灾害的实时立体监测,并充分挖掘采集的数据信息,进行了数据处理和模型模拟,实现了蓝藻灾害信息实时接收、快速精确的处理和及时发布。

Description

一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法 技术领域
本发明属于环境监测评价和数据挖掘技术领域,特别是涉及面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法。
背景技术
湖泊富营养化引起的一个普遍现象是:在适宜的水文气象条件下,许多浮游植物,尤其是那些具有浮力或运动能力的藻类,会发生异常增殖,并聚集形成表面水华,进而引发一系列严重的生态环境问题,包括:水体透明度下降,溶解氧减少,水生动植物死亡,水体散发异味,生物多样性下降,通过食物链损害人类健康等。这种由富营养化和蓝藻异常增殖引起的生态灾害在全球范围内的持续扩张已经对许多著名的湖泊生态系统的生态健康和可持续发展构成了巨大的威胁,包括:美国伊利湖,加拿大温尼伯湖,欧洲的波罗的海,非洲维多利亚湖及日本的琵琶湖等。不过,尽管湖泊蓝藻灾害非常的严重,但是直到目前还没有办法彻底解决此问题。在此前提下,做好蓝藻灾害监测和预防工作是减轻此生态灾害,减少社会经济损失的关键。
目前,已经有不少有关蓝藻监测的技术方法研究报道。中国专利申请201410023795.8提出的“一种大型浅水湖泊蓝藻水华MODIS卫星高精度监测方法”是利用遥感技术来监测蓝藻水华。在没有云层覆盖影响下,这种方法最多只能提供每天一次数据。如果遇到云层覆盖,则不能提供有效数据。但是,研究已证明水华的形成和消失过程可以仅持续数个小时。而遥感技术的采样频率显然不能准确反映水华的快速的动态实时变化。相似地,中国专利申请201020219363.1提出的“用于蓝藻监测的浮标”实时监测虽然能够以高时间分辨率记录与蓝藻灾害相关的环境因子变化过程,但是该技术仅能提供设备所在点的水环境信息,这显然也不能有效记录呈现空间高度分异的蓝藻水华面状信息。杨宏伟等公开了一种“基于物联网技术的太湖蓝藻水华预警平台”,该平台基于多源数据,根据物联网的四层内涵设计,能够实现未来3天蓝藻水华预警区域的发生概率的预测,并通过Internet发布信息;但从其设计框架及文字说明来看,该技术存在主要问题为:(1)并未包括卫星接收天线,其获取的数据并非实时的卫星遥感数据,因此该平台不能实现自动实时获取卫星遥感数据的功能;(2)自动监测站仅详细描述了叶绿素传感器的设置,对其他仪器设备的监测指标、布置方式和参数设定并未提及,也没有介绍实时监控、供电和安全等功能;(3)该技术仅提及了蓝藻生物量模型,并不具有营养盐循环、沉积物侵蚀悬浮和溶解氧动态等与藻类生命过程密切相关的生态过程的模拟功能;(4)该技术还存在不能充分发掘监测产生的数据价值的缺点。遥感和自动监测均可以产生海量的有关蓝藻灾害的数据。但是现有技术并不能充分利用这部分数据为湖泊研究和管理服务,更不能为普通公众提供有价值的、可读性强的蓝藻灾害信息。因此,为了更加及时准确的获得蓝藻灾害信息,并充分提升这些信息的价值,既需要测控系统采集时空连续的与蓝藻相关数据,也需要具有高效的数据分析和处理技术和方法,包括数据统计分析、数据同化、模型预测、参数的实时校正、灾害评估和灾害信息发布等。只有建立蓝藻灾害立体监控和数据挖掘系统和方法,才能满足经济社会发展和生态环境保护对蓝藻灾害监测和预防工作的需求。
发明内容
本发明目的在于克服现有技术的不足,提供一种面向蓝藻灾害的立体监控及数据挖掘系统和方法。采用本发明提供的系统和方法,针对湖泊蓝藻水华时空变化大的问题,通过集成卫星遥感、自动监测和人工巡测技术收集实时蓝藻灾害数据,并利用数据库、数据同化和数值模型 等方法开展数据挖掘,实现自动、实时和准确的采集、处理和提取蓝藻灾害信息,为快速、及时和准确的应对蓝藻灾害提供有价值的和辨识度高的监测和预防信息,以便能最终实现将灾害引起社会经济和生态环境损失降低到最小的根本目标。
为实现上述技术目的,本发明采用如下技术方案:
一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统,包括监测子系统和数据处理子系统;
所述监测子系统用于采集待监测数据;包括利用遥感进行监测的遥感监测子系统、利用自动监测站进行监测的自动监测子系统和人工采集待监测数据的人工巡测子系统;
所述遥感监测子系统包括卫星数据接收天线、计算机和供电系统,所述供电系统用于为耗电装置供电,卫星数据接收天线接收卫星数据,并向计算机传输,通过计算机进行遥感反演处理后,将处理后的数据通过互联网传输至数据处理子系统;
所述自动监测子系统为多个自动监测站通过无线网络连接形成的监测网络,单个自动监测站由水面支撑系统、供电系统、安全警示系统和数据采集系统组成;所述水面支撑系统为自动监测站硬件装置的承重结构;所述供电系统用于为自动监测站的耗电装置供电;所述安全警示系统用于安全警示,防止自动监测站遭受意外破坏;所述数据采集系统用于采集包括从气象仪器、水文仪器、水质仪器和视频仪器中获取的待监测数据,并将采集的数据通过无线网络传输到数据处理子系统;
水面支撑系统是整个系统的承重结构,可以是浮标体、栈桥或水上平台。支撑系统上部出露于水面,下部固定于湖底,起到支撑整个系统结构的功能。供电系统固定在水面支撑系统上,通过电缆为其他耗电设备提供电力。安全警示系统固定在水面支撑系统之上,包括航标灯、荧光带和警示标语等。
所述人工巡测子系统为人工采集数据,并将数据通过互联网传输至数据处理子系统;
所述数据处理子系统用于接收和处理监测子系统获取的数据;包括服务器、阵列机、计算机、计算工作站、硬件防火墙、路由器、网线和供电设备;服务器采用双机热备模式,两台服务器和用于数据存储的阵列机通过三叉电缆实现心跳连接,所有计算机、服务器和计算工作站均通过网线与硬件防火墙相连,硬件防火墙通过连接路由器与外网连通;
本发明中,服务器采用双机热备的模式,将中心服务器安装成互为备份的两台服务器,并且在同一时间内只有一台服务器运行。当其中运行着的一台服务器出现故障无法启动时,另一台备份服务器会迅速的自动启动并运行。计算工作站主要用于三维数值模型的运算。硬件防火墙主要用于保护整个系统免受攻击,路由器则是为整个系统提供网络服务。上述这些设备将被放置在机柜中,机柜电源与不间断供电电源相连接。正常情况下,所有设备由外接电源供电。在外接电源意外间断时,由不间断供电电源提供整个系统的电力供应。
数据处理子系统接收数据后,对接收的数据进行数据备份和数据处理,获取湖泊蓝藻灾害的预测指标和风险评估,并通过公共平台发布。
本发明还提供了一种面向蓝藻灾害的立体监控和数据挖掘方法,具体包括如下步骤:
(1)通过遥感监测、自动监测和人工巡测三种途径获取与待监测湖泊蓝藻灾害相关的监测指标数据,包括气象指标、水文指标、水质指标和视频影像,获取的数据通过互联网传输至数据中心;
其中,所述遥感监测指通过卫星遥感实时监测;遥感监测数据采集由卫星接收天线接收卫星信号,并传送到遥感处理计算机中,由卫星数据反演软件生成各项遥感指标,此后,这些数据通过局域网内计算机共享功能实现向服务器数据中心发送;
所述自动监测指由多个自动监测站通过无线网络连接形成监测网络,对待监测指标进行监测;自动监测则通过配置的各类传感器收集气象、水文和水质指标及视频,并由通信模块通过商用通信网络向服务器数据中心发送,并由安装在服务器端的商业软件完成数据接收;
所述人工巡测指通过人工方式进行指标检测;人工巡测是指在规定的时间内通过对预先设 定的监测站点开展现场数据和水样采集,并将水样送到实验室检测相关指标,人工巡测数据由人工通过网站上载到服务器数据中心;
传送至服务器的遥感监测、自动监测和人工巡测数据以特定的文件命名方式存储在服务器数据中心中,以方便各类数据处理程序对原始数据的识别;
其中,适用遥感监测指标体系确定方式为:基于水温和水色等水质指标对卫星携带的光学传感器发射和接收光谱的影响,选取既与蓝藻灾害密切相关的,也能利用反演算法从光谱信号中提取有效信息的水质指标作为卫星反演指标,包括:水温、透明度、悬浮物质(SS)、叶绿素a、蓝藻水华面积和强度等。遥感监测时间分辨率一天以内,空间分辨率在1km以内;
适用自动监测指标体系确定方式为:基于对市场上气象、水文和水质传感器技术成熟度调研,选取既与蓝藻灾害密切相关的,又具有较高成熟度的传感器技术支持的指标作为自动监测指标,包括:气象指标(风速、风向、气压、温度、湿度、太阳辐射和降雨量等);水文指标(三维流速剖面、水深、有效波高和周期等);水质指标(水温、溶解氧、浊度、电导率、氧化还原电位、藻蓝素、叶绿素等);视频影像;
适用人工巡测指标体系确定方式为:不能使用遥感反演和自动监测的指标,而又与蓝藻灾害密切相关的指标,将通过人工巡测的方式获取,包括:总氮、溶解性总氮、氨氮、硝态氮、亚硝态氮、总磷、溶解性总磷、正磷酸盐和其它蓝藻衍生污染物等。
遥感监测、自动监测和人工巡测数据以特定的文件命名方式存储在服务器中,以方便各类数据处理程序对原始数据的识别。
(2)数据中心对接收的数据进行数据备份和数据预处理,所述数据预处理包括如下步骤:
根据设定的数据采集时间间隔对数据进行检查,如果数据有中断,则通过时间插值方法对数据进行插值处理;
对遥感监测获取的卫星数据,通过空间插值填补云层覆盖区域的数据缺失;
对数据进行异常判定和处理,对于判定的异常数据,采用异常数据的前一个数据代替该异常数据;
数据经预处理后,连同数据中心接收的原始数据一同传输至数据库存储;
(3)根据数据库中存储的数据源构建待监测湖泊的三维数值模型,具体为:
构建待监测湖泊的水动力模型;
在待监测湖泊的水动力模型上叠加物质迁移转化模型,两种模型的方程耦合计算;其中,所述物质迁移转化模型计算的标量包括光照、悬移质、藻类生长、营养盐循环和溶解氧;
采用有限差分求解模型,获取数值模型模拟数据;
采用的三维数值模型可以是FVCOM、ELCOM-CAEDYM和EFDC等现有的模型,也可以是基于环境流体力学理论自主开发的模型;
(4)根据数值模型模拟数据,进行待监测湖泊蓝藻灾害风险评估;灾害风险评估可采用现有的成熟算法;
(5)在公共平台上展示湖泊蓝藻灾害预测预警信息。
本发明的方法,进一步的,所述步骤(2)中,对数据库中存储的数据进行分类存储,具体如下:
对于单点时间连续的数据,以单个数据表存储单个监测站的所有数据;
对于二维数据直接存放在数据表中;
对于三维数值模型生成的数据,以时间为节点,存放在数据表中;
对于图像或视频数据,将图像或视频存储在阵列机中,在数据库中建立数据表记录图像或视频的路径,字段包括编号、时间和图像/视频路径,以索引的方式访问图像或视频。将遥感监测、自动监测和人工巡测数据以特定的文件命名方式存储在服务器中,便与各类数据处理程序对原始数据的识别。
其中,所述数据库面向多源异构数据集构建,此处选用一种商用软件,如微软公司的SQL Server、Access和甲骨文公司的Oracle等;优选采用Oracle。
本发明的方法,所述步骤(2)中,所述时间插值方法采用线性插值、样条函数插值或分段插值;
空间插值方法采用邻近点法、克里格法或反距离加权法;
异常值判定和处理的依据为趋势检验、专家经验或数值比对。
优选的,所述时间插值方法采用线性插值,算法具体如下:
对于数据集中a1和a2两个数据,依据时间顺序,中间缺少b1,b2,……,bn数据,那么:
Figure PCTCN2017089012-appb-000001
其中i=[1,n]。
所述空间插值方法采用反距离加权法,算法具体如下:
假设空间点坐标(x0,y0)处指标值缺测C(x0,y0),设定一个搜索半径,使得在此半径范围内至少包含3个数据点;然后使用这些已知数据点数据求取未知点的指标值:
Figure PCTCN2017089012-appb-000002
式中,C(x1,y1),C(x2,y2),…,C(xn,yn)分别表示括号内坐标点处实测指标值;d1,d2,…,dn分别表示括号内坐标点到空间点坐标(x0,y0)对应的直线距离,n≥3。
采用5倍方差法进行异常值判定,具体方法如下:将第m个实测数据am及其前后5个数据求平均值和方差:
Figure PCTCN2017089012-appb-000003
Figure PCTCN2017089012-appb-000004
Figure PCTCN2017089012-appb-000005
为判断区间,满足
Figure PCTCN2017089012-appb-000006
的数据则为常规数据,否则以第m-1个数据代替am
本发明的方法,所述步骤(3)中,采用的三维数值模型可以是FVCOM、ELCOM-CAEDYM和EFDC等现有的模型,也可以是基于环境流体力学理论自主开发的模型;本发明优选提供了一种自主开发的模型,具体如下:
水动力模型的控制方程如下所示:
Figure PCTCN2017089012-appb-000007
Figure PCTCN2017089012-appb-000008
Figure PCTCN2017089012-appb-000009
Figure PCTCN2017089012-appb-000010
Figure PCTCN2017089012-appb-000011
式中:u,v,w分别为x,y,σ三个方向上流速;H和t为水深和时间;g和ξ为重力加速度和水位;f为科里奥利力,P为压力,Bx,By和BT分别为x向动量,y向动量和温度方程的由坐标转换而引入的小项;T为温度;Sh为外部进入系统的热量;Cp为水体的热容量;Kh,Kv为热量在水平及垂直方向上的扩散系数;ρ为水体密度,ρ0为水体参考密度;Ah为水平涡粘性系数;Av为垂向涡粘系数,采用下式定义:
Figure PCTCN2017089012-appb-000012
Figure PCTCN2017089012-appb-000013
式中:v0、m0、m1值分别为5.0×10-6m2/s,0.1和-1;l为普朗特长度;Ri为Richardson数,反映流体稳定性状况,其表达式为:
Figure PCTCN2017089012-appb-000014
在水气界面σ=1的风能输入和水土界面σ=0的摩阻力可分别表达为:
Figure PCTCN2017089012-appb-000015
Figure PCTCN2017089012-appb-000016
式中:ρa和ρs分别表示空气密度和表层水体密度;CWD是风拖拽系数;WS是水表以上10m高度处风速;CSD为湖底摩擦系数;
采用分裂算子技术求解水动力方程(5)、(6)和(7),生成内外两种模态;然后利用有限差分离散内外两种模态,其中水平和时间差分格式为显式,垂直差分格式为隐式,并采用低通滤波器对水面位移在时间方向上进行平滑处理;最后采用追赶法求解超大型稀疏矩阵。
在水动力模型上叠加的物质迁移转化模型的控制方程如下所示:
γPAR(I,J,K)=γ0,PAR1,PARCHLA(I,J,K)+γ2,PARSED(I,J,K)         (15)
Figure PCTCN2017089012-appb-000017
Figure PCTCN2017089012-appb-000018
式中:γPAR(I,J,K)、γ0,PAR分别表示总衰减系数和纯水衰减系数;γ1,PAR、γ2,PAR分别表示藻类比衰减系数和非藻类颗粒物比衰减系数;CHLA(I,J,K)为以叶绿素a浓度表示的浮游植物 生物量;SED(I,J,K)非藻类颗粒物浓度;u,v,w为新时间步的三维流速;S为悬移质浓度;ws为悬浮物沉降速率;J0为水土界面通量项,包括侵蚀通量和沉降通量;Ci表示第i种物质浓度;SKi表示生化过程项,i=1,2,3,4,5分别表示叶绿素a、浮游动物、可利用磷、可利用氮、溶解氧及五日生化需氧量;
基于水动力模型计算结果,利用有限差分法在矩形网格中离散上述方程:空间上离散采用迎风格式;水平和时间差分格式为显式,垂直差分格式为隐式;追赶法求解超大型稀疏矩阵;
方程(17)涉及的生化项采用下列方法计算:
藻类生化项:
Figure PCTCN2017089012-appb-000019
式中:μ为总生长率;KM为非牧食死亡率;KS为上浮率;ZP为浮游动物滤食率;CHLAi,j,k表示i,j,k网格的浮游植物生物量;此处的τa、τf、τp及下述公式涉及的τZP,τKB,τs,τd,τso均为温度影响因子,表达式分别为θa T-20,θf T-20,θp T-20,θZP T-20,θKB T-20,θs T-20,θd T-20及θso T-20,其中θa,θp,θZP,θKB,θs,θd和θso为温度乘子;
浮游动物生化项:
Figure PCTCN2017089012-appb-000020
式中:μzp为浮游动物生长率;KCHLA为浮游动物牧食半饱和参数;BFISH和FISH分别表示鱼滤水率和鱼生物量;ZOOP表示浮游动物生物量;
DTP和DTN生化项:
Figure PCTCN2017089012-appb-000021
Figure PCTCN2017089012-appb-000022
式中:FMRP、FMRN、FMDP和FMDN分别为藻类代谢物磷转化率、代谢物氮转化率、死亡残骸磷转化率和死亡残骸氮转化率;KD和KM分别为藻类代谢率和死亡率;RPJ和RNJ分别沉积物磷和氮静态释放率;RPD和RND分别沉积物磷和氮动态释放率;SEDF水土界面悬浮物通量;ZDP和ZDN分别藻类对磷和氮吸收率;KPS和KNS为DTP和DTN的沉降率;
DO和BOD生化项:
Figure PCTCN2017089012-appb-000023
SK6=(τNP·ZHD+τR·ZHR)CHLA-τKB·KB·BOD            (23)
式中:KOD为大气复氧率;DOSAT为饱和DO;Hs为特征波高;PP藻类光合产氧;ZHY为藻类呼吸耗氧;RO为沉积物耗氧率;ZHD和ZHR为藻类死亡和代谢物BOD产生率;KB为BOD降解速率。
所述步骤(3)中,三维数值模型从数据库中读取某时刻点的初始化数据和边界条件数据,以提高计算速度和效率,并可依据上个模拟期内的实测数据,采用Monte Carlo法对模型参数组合进行优化,以提高本轮模拟期模型预测精度;在完成所需数据输入后,模型自动激活开展本轮模拟运算,并将本轮计算结果回传给数据库;在结束上述过程后,模型转入休眠状态,等待下轮模拟运算。这样可实现自主化滚动预报。
所述步骤(4)中,蓝藻灾害评估算法作用是对将抽象的数据转化为易为公众理解的文字或图像,有助于提高预警效率。基于蓝藻灾害相关的理论研究,目前评估算方法有专家评分体系、卡尔森营养状态指数和蓝藻灾害发生概率法等。基于这些方法,利用程序设计语言编制算法程序,并将之安装在服务器中。这种程序能够自动检索数据库中最新的立体监控数据和模型预测数据,并对数据进行计算分析,生成蓝藻灾害风险图,并以网格化采样方式对风险图进行 简单的数理统计,给出描述性结论。这些图表和结论最终将通过网络向公众发布。
所述步骤(5)中,在公共平台上展示湖泊蓝藻灾害预测预警信息,所述的公共平台可选用基于互联网的软件平台,利用相关网站进行信息发布,蓝藻灾害信息发布网站是基于互联网的软件平台,是管理立体监控及数据挖掘系统的对话窗口,通过网站形式实现用户与系统之间的信息交互。本发明构建的蓝藻灾害信息发布平台可具有以下功能:1)提供实用的蓝藻灾害资讯;2)能够展示立体监控系统所获得的蓝藻灾害实时信息;3)能够以表、图和电子地图相结合的展示功能;4)能够检索、统计和下载历史数据;5)用户能够操控三维数值模型;6)用户能够干预蓝藻预警文件的制作和发布;7)利用多种途径自动向相关单位和个人发布最新的监测和预测信息;8)具有绘制等值线图和制作简单的Flash动画功能;9)具有系统管理功能。
本发明的优点:
(1)面向蓝藻灾害的监测指标选择的科学性,本发明以蓝藻灾害发生、发展和结束等整个过程的研究为基础,经过大量的实验和调查,确定了所有与蓝藻灾害过程密切相关的指标,并对各指标采集技术上的可行性进行分析;
(2)通过将遥感监测、自动监测和人工监测相结合,实现对蓝藻灾害的实时立体监测。其中,遥感监测可以采集面状分布的指标,自动监测可以极高的频率采集时间连续的指标,人工监测则可以弥补通过上述两种技术不能采集的指标。
(3)研发了数据自动接收和预处理程序。这些程序不但可以自动将采集的数据存储入数据库,还能够对原数据进行预处理,以剔除异常数据,或者对缺测数据进行时空插补,使得数据库不仅存储原始数据,还有一套经过预处理的时空连续数据,用于进一步的数据挖掘。
(4)采用自主设计的三维数值模型,考虑了营养盐循环、沉积物侵蚀悬浮和溶解氧动态等与藻类生命过程密切相关的生态过程的模拟功能,计算结果更准确。
(5)建立了安全、稳定和智能化数据中心。数据中心的双机热备、UPS电源保护体系和网络硬件防火墙等技术可以使得整套系统能够始终安全稳定的运行。此外,数据库、数据接收处理、三维数值模型、参数优化程序和蓝藻灾害信息发布网站等软件的配置,不仅能够自动接收、处理和挖掘立体监控数据,还能够自主生成预警信息和通过互联网向相关部门和个人发布。实现了蓝藻灾害信息实时接收、快速处理和及时发布。
(6)面向蓝藻灾害的三维数值模拟技术和蓝藻灾害评估方法与立体监控系统的无缝耦合。尽管已经有不少蓝藻灾害监测系统被报道出来,但是这些监控系统只实现了数据的收集功能,却不知道如何利用收集到的海量数据。本发明将数值模型和蓝藻灾害评估方法植入到数据中心,并将其与立体监控系统无缝耦合,通过专业性的计算和分析,最大化提取海量监测数据中的有价值信息,为社会经济发展和居民生产生活服务。
下面结合具体实施例对本发明进行详细描述。本发明的保护范围并不以具体实施方式为限,而是由权利要求加以限定。
附图说明
图1是本发明方法流程示意图;
图2是本发明实施例1自动监测站分布图;
图3是本发明实施例1模型数值计算网格划分示意图。
具体实施方式
下面结合附图说明和具体实施方式对本发明的技术方案作进一步描述。以下实施例用于说明本发明,但不用来限制本发明的范围。
实施例1
本实施例以太湖为例,对本发明的方法作进一步描述。
图1所示为本发明方法的流程图,本发明的面向湖泊蓝藻灾害的立体监控及数据挖掘方法包括如下步骤:
(1)通过遥感监测、自动监测和人工巡测三种途径获取与待监测湖泊蓝藻灾害相关的监测指标数据,包括气象指标、水文指标、水质指标和视频影像,获取的数据通过互联网传输至数据中心;
其中,所述遥感监测指通过卫星遥感实时监测;
所述自动监测指由多个自动监测站通过无线网络连接形成监测网络,对待监测指标进行实时监测;
所述人工巡测指通过人工方式进行指标检测;
位于中国长江三角洲的太湖,是中国第三大淡水湖,目前正面临严重的富营养化和蓝藻水华威胁。2007年发生在无锡的水危机事件,对以太湖为水源地的人民群众生产和生活造成巨大损失。依据步骤(1)所述方法,对所有与蓝藻水华灾害密切相关的指标开展现场和实验室对比测试,并充分调研相关传感器技术发展情况,给出各指标实现遥感监测、自动监测和人工巡测的可行性和精度,如表1所示,为最大程度的减少人力并提高精度,采用如下的测量指标分配构建太湖蓝藻水华灾害监测指标体系:遥感监测指标包括:有效波高、水温、浊度、透明度、叶绿素a、水华面积、水华强度;自动监测指标包括:风速、风向、气压、温度、湿度、太阳辐射和降雨量,三维流速剖面、水深、有效波高和周期,水温、溶解氧、浊度、电导率、氧化还原电位、藻蓝素、叶绿素,视频影像;人工巡测指标包括:各类氮磷浓度、叶绿素a、浮游植物、有毒有害物、藻毒素和底栖动物。
表1 面向太湖蓝藻水华灾害的监测指标监测可行性和精度调查
Figure PCTCN2017089012-appb-000024
Figure PCTCN2017089012-appb-000025
(2)数据中心在收到原始数据后,接收的数据进行数据备份,并通过自主编制的数据预处理程序,对数据缺测和异常值等情况开展检查,进行数据预处理,对所述数据预处理包括如下步骤:
根据设定的数据采集时间间隔对数据进行检查,如果数据有中断,则通过时间插值方法对数据进行插值处理;所述时间插值方法采用线性插值、样条函数插值或分段插值;本实施例中采用
对遥感线性插值,算法具体如下:
对于数据集中a1和a2两个数据,依据时间顺序,中间缺少b1,b2,……,bn数据,那么:
Figure PCTCN2017089012-appb-000026
其中i=[1,n]。
对于涉及的卫星反演数据,在有云层遮盖的情况下,某些区域不能得到有效的反演结果,此时使用空间插值来填补云层覆盖区域的数据缺失,空间插值方法可采用邻近点法、克里格法或反距离加权法;本实施例中使用反距离加权法实现空间插值。假设空间点坐标(x0,y0)处指标值缺测C(x0,y0),设定一个搜索半径,使得在此半径范围内至少包含3个数据点;然后使用这些已知数据点数据求取未知点的指标值:
Figure PCTCN2017089012-appb-000027
式中,C(x1,y1),C(x2,y2),…,C(xn,yn)分别表示括号内坐标点处实测指标值;d1,d2,…,dn分别表示括号内坐标点到空间点坐标(x0,y0)对应的直线距离,n≥3。
异常值判定和处理的依据为趋势检验、专家经验或数值比对,对于判定的异常数据,采用异常数据的前一个数据代替该异常数据;本实施例中使用5倍方差法对数据进行异常判定和处理,将第m个实测数据am及其前后5个数据求平均值和方差:
Figure PCTCN2017089012-appb-000028
Figure PCTCN2017089012-appb-000029
Figure PCTCN2017089012-appb-000030
为判断区间,满足
Figure PCTCN2017089012-appb-000031
的数据则为常规数据,否则以第m-1个数据代替am
数据经预处理后,连同数据中心接收的原始数据一同传输至数据库存储;
本实施例的数据库采用甲骨文公司的Oracle数据库软件,不同来源的数据具有不同的数据结构,因此在Oracle数据库中采取不同的存储策略:
对于单点时间连续的数据,以单个数据表存储单个监测站的所有数据;
对于二维数据直接存放在数据表中;
对于三维数值模型生成的数据,以时间为节点,存放在数据表中;
对于图像或视频数据,将图像或视频存储在阵列机中,在数据库中建立数据表记录图像或视频的路径,字段包括编号、时间和图像/视频路径,以索引的方式访问图像或视频。
具体的,例如:对于通过自动监测和人工巡测等这类单点时间连续的数据,以单个数据表存储单个站的所有数据。数据表名称为监测站名称;数据表字段为指标名称;数据记录为以时间顺序存储的实测数据值。例如,如图2所示的自动监测站EMB16水温、叶绿素a和水深等数据的存储过程为:1)先建立一个以EMB16命名的数据表;2)表的字段名称为:编号,时间,水温,叶绿素a和水深等;3)表的第一行则存储EMB16记录到的第一条数据。
对面状的卫星遥感数据采取两种存储策略:卫星图片仅在数据表中存放影像存放的路径;由卫星图片反演得到的二维数据则直接存放在数据表中。卫星图片存储在阵列机中以“卫星图片”命名的文件夹中,在数据库中,建立数据表,并将其命名为“卫星图片路径”;字段包括:编号,时间,图片路径,备注;数据表中每条记录对应某个时间采集的卫星图片。卫星反演之后,通常每个像元将对应一个数据。例如:一张南北跨度250*M米的,东西跨度250*N米,分辨率为250m的卫星图片,对其水温指标反演之后将生成一个具有M行N列的数组。那么该数组存储方式为:建立数据表,并将其命名为“反演水温”;字段包括:编号,水层,时间,水温1,水温2,…….,水温N;表的第一行存储数组的第一行的第1,2,……,N列数据,表的第二行存储数组的第二行的第1,2,……,N列数据,以此类推,直到完成整个二维数组的存储;在完成由第一个时刻点卫星图片反演的水温数据后,紧接着存储第二个时刻点的卫星反演数值,以此类推,其中水层和时间可以标记不同时刻点的卫星反演数据。
三维数值模型生成的数据,此将是一个三维数组。例如将研究湖泊水平方向划分为M行和N列,水深方向划分为K层,则对于模拟得到的水温,则是TEMP(M,N,K)。与“反演水温”数据存储类似,对于三维数组的存储为在一个时刻点实现K个二维数组的存储:数据表命名为“模拟水温”;字段包括:编号,水层,时间,模拟水温1,模拟水温2,…….,模拟水温N;每一个水层的存储方式与反演水温一致;在完成第一水层二维数组存储后,紧接着存储第二水层的二维数值,以此类推,直到完成K层二维数值存储。完成第一个时间点三维数组存储后,再进行第二个时间点三维数值的存储,以此类推,其中水层和时间可以标记同水层和不同时刻点的模型模拟数据。
(3)根据数据库中存储的数据源构建待监测湖泊的三维数值模型,具体为:
构建待监测湖泊的水动力模型;
在待监测湖泊的水动力模型上叠加物质迁移转化模型,两种模型的方程耦合计算;其中,所述物质迁移转化模型计算的标量包括光照、悬移质、藻类生长、营养盐循环和溶解氧;
采用有限差分求解模型,获取数值模型模拟数据;
本实施例中采用自主构建的模型,描述湖水运动的方程组如下:
Figure PCTCN2017089012-appb-000032
Figure PCTCN2017089012-appb-000033
Figure PCTCN2017089012-appb-000034
Figure PCTCN2017089012-appb-000035
Figure PCTCN2017089012-appb-000036
式中:u,v,w分别为x,y,σ三个方向上流速;H和t为水深和时间;g和ξ为重力加速度(9.8m/s)和水位(m);f为科里奥利力(=2Ωsin(φ)(u,v)),P为压力,Bx,By和BT分别为x向动量,y向动量和温度方程的由坐标转换而引入的小项;T为温度;Kh,Kv为热量在水平及垂直方向上的扩散系数;Sh为外部进入系统的热量;Cp为水体的热容量;ρ为水体密度,ρ0为水体参考密度;Ah为水平涡粘性系数,太湖取5m2/s;Av为垂向涡粘系数,采用下式定义:
Figure PCTCN2017089012-appb-000037
Figure PCTCN2017089012-appb-000038
式中:v0、m0、m1值分别为5.0×10-6m2/s,0.1和-1;l为普朗特长度;Ri为Richardson数,反映的是流体稳定性状况,其表达式为:
Figure PCTCN2017089012-appb-000039
在水气界面(σ=1)的风能输入和水土界面(σ=0)的摩阻力可分别表达为:
Figure PCTCN2017089012-appb-000040
Figure PCTCN2017089012-appb-000041
式中:ρa和ρs分别表示空气和表层水体密度(=1000kg/m和1.3kg/m);CWD为风拖拽系数,此处取为0.001;WS为水表以上10m高度处风速(m/s);CSD为湖底摩擦系数,取0.003。
采用分裂算子技术求解水动力方程(5)、(6)和(7),生成内外两种模态;然后利用有限差分离散内外两种模态,其中水平和时间差分格式为显式,垂直差分格式为隐式,并采用低通滤波器对水面位移在时间方向上进行了平滑处理;最后采用追赶法求解超大型稀疏矩阵。在本实施例中,计算网格设置为:在水平方向上采用边长为1000m的矩形网格将计算域分成69×69个网格;垂直方向上分为5层。时间步长取30s。
在上述水动力模型基础上,再叠加光照、悬移质、藻类生长、营养盐循环和溶解氧等标量物质迁移转化模型:
γPAR(I,J,K)=γ0,PAR1,PARCHLA(I,J,K)+γ2,PARSED(I,J,K)        (15)
Figure PCTCN2017089012-appb-000042
Figure PCTCN2017089012-appb-000043
式中:γPAR(I,J,K)、γ0,PAR分别表示总衰减系数和纯水衰减系数;γ1,PAR、γ2,PAR分别表示藻类比衰减系数和非藻类颗粒物比衰减系数;CHLA(I,J,K)以叶绿素a浓度表示的浮游植物生物量;SED(I,J,K)非藻类颗粒物浓度。u,v,w为新时间步的三维流速;S为悬移质浓度;ws悬浮物沉降速率;J0水土界面通量项,包括侵蚀通量和沉降通量;Ci表示第i种物质浓度;SKi表示生化过程项,i=1,2,3,4,5分别表示叶绿素a、浮游动物(ZOOP)、可利用磷(DTP)、可利用氮(DTN)、溶解氧(DO)及五日生化需氧量(BOD)。
基于水动力模型计算结果,利用有限差分法在矩形网格中离散上述方程:空间上离散采用迎风格式;水平和时间差分格式为显式,垂直差分格式为隐式;追赶法求解超大型稀疏矩阵。此外,方程(17)涉及的生化项采用下列方法计算。
藻类生化项:
Figure PCTCN2017089012-appb-000044
式中:μ为总生长率;KM为非牧食死亡率;KS为上浮率;ZP为浮游动物滤食率;CHLAi,j,k表示i,j,k网格的浮游植物生物量(此处以叶绿素a的浓度表示);此处的τa、τf、τp及下文的τZP,τKB,τs,τd,τso均为温度影响因子,表达式分别为θa T-20,θf T-20,θp T-20,θZP T-20,θKB T-20,θs T-20,θd T-20及θso T-20,其中θa,θp,θZP,θKB,θs,θd和θso为温度乘子。
浮游动物生化项:
Figure PCTCN2017089012-appb-000045
式中:μzp浮游动物生长率;KCHLA浮游动物牧食半饱和参数;BFISH和FISH分别鱼滤水率和鱼生物量;ZOOP表示浮游动物生物量;
DTP和DTN生化项:
Figure PCTCN2017089012-appb-000046
Figure PCTCN2017089012-appb-000047
式中:FMRP、FMRN、FMDP和FMDN分别为藻类代谢物磷转化率、代谢物氮转化率、死亡残骸磷转化率和死亡残骸氮转化率;KD和KM分别为藻类代谢率和死亡率;RPJ和RNJ分别沉积物磷和氮静态释放率;RPD和RND分别为沉积物磷和氮动态释放率;SEDF为水土界面悬浮物通量;ZDP和ZDN分别表示藻类对磷和氮吸收率;KPS和KNS为DTP和DTN的沉降率。
DO和BOD生化项:
Figure PCTCN2017089012-appb-000048
SK6=(τNP·ZHD+τR·ZHR)CHLA-τKB·KB·BOD           (23)
式中:KOD为大气复氧率;DOSAT为饱和DO;Hs为特征波高;PP藻类光合产氧;ZHY为藻类呼吸耗氧;RO为沉积物耗氧率;ZHD和ZHR为藻类死亡和代谢物BOD产生率;KB为BOD降解速率。
三维数值模型从数据库中读取某时刻点的初始化数据和边界条件数据,并依据上个模拟期内的实测数据,采用Monte Carlo法对模型参数组合进行优化;在完成所需数据输入后,模型自动激活开展本轮模拟运算,并将本轮计算结果回传给数据库;在结束上述过程后,模型转入 休眠状态,等待下轮模拟运算。
本实施例汇总,面向太湖蓝藻水华的三维数值模型计算采用矩形网格。根据对太湖蓝藻水华易发水域地形的现场测定,水平面上将全湖划分为包括陆地和水域的4900个计算网格,如图3所示。基于此网格图,模型在预报日12:00从Oracle数据库中读取湖泊风浪、湖流、水温、光照、叶绿素a、营养盐、溶解氧和有机物的实测值,并运用反距离权重插值法将这些数据插值到4900个计算网格中(陆地网格内值设为―9999),以此作为本轮预报的初始化浓度场。同时,依据上一轮预报期内(3d)实测的水环境数据,对本轮预报的模型参数进行优化,采用Monte Carlo法实现模型参数组合的优选。在完成初始化和参数优化后,还需要获取本轮预报期3d的天气状况。模型所需的未来3d的风速、风向、气温、辐射、降水、气压和相对湿度由天气观测与预报模型(WRF)计算。最后,模型自动激活对未来3d的湖泊风浪、湖流、水温、光照、叶绿素a、营养盐、溶解氧和有机物变化开展模拟预测,并将生成的三维数组传回Oracle数据库。
(4)根据数值模型模拟数据,进行待监测湖泊蓝藻灾害风险评估;
蓝藻灾害评估算法作用是对将源自数据库的抽象的数据转化为易为公众理解的文字或图像。本实施例中,采用专家评估体系的太湖蓝藻水华风险评估算法,利用Intel公司的Visual Fortran和微软公司的Visual Studio,研发评估算法程序,该程序能够自动从数据库中读取三维数值模型模拟数据,并对未来3d蓝藻水华风险做出评估。评估结果在网站上发布。
基于专家评估体系的太湖蓝藻水华风险评估算法是依据三维数值模型模拟的未来3d的湖泊风浪、湖流、水温、光照、叶绿素a、营养盐、溶解氧和有机物变化空间分布,引入专家打分,判定太湖不同时间,不同地点出现蓝藻水华灾害的风险等级,并以不同的颜色表征风险等级。具体实现方式如下:
(1)对太湖进行水域功能划分。基于自然条件相似性、湖泊污染现状相似性、使用目标的相似性、行政管理完整性的划分原则,以自然条件指标(水深、湖流特性、生物)、水质污染指标(TP、TN、CODMn、有机污染)、使用目标指标(饮用、渔业、游览、自然保护区)和行政区划为指标体系,对水域进行功能划分。
(2)建立了蓝藻灾害风险评估指标体系和分级体系,基于科学性、代表性、可定量、可达性原则,将蓝藻灾害生态风险按照重要性分为三个层次:关键性指标、重要指标和一般指标。
(3)据专家经验,对不同水体单元和水环境指标赋不同权重,然后依据下式计算蓝藻灾害生态风险分值:
Figure PCTCN2017089012-appb-000049
式中:Yi为指标i的打分结果;Wi为指标i的权重,Ci为水体单元权重,G为综合评分。依据此计算结果和专家经验,以不同颜色表示各个功能水域蓝藻灾害的不同风险等级,包括:极重、重灾、中灾、轻灾和无灾等。评估算法的具体实现方式可参照刘聚涛等人的文献(太湖蓝藻水华灾害风险分区评估方法研究,中国环境科学,2011,31(3):498-503),本发明中不再赘述。
(5)在公共平台上展示湖泊蓝藻灾害预测预警信息。
在公共平台上展示湖泊蓝藻灾害预测预警信息,所述的公共平台可选用基于互联网的软件平台,利用相关网站进行信息发布,蓝藻灾害信息发布网站是基于互联网的软件平台,是管理立体监控及数据挖掘系统的对话窗口,通过网站形式实现用户与系统之间的信息交互。本实施例构建的蓝藻灾害信息发布网站是基于互联网的软件平台,是管理立体监控及数据挖掘系统的对话窗口,通过网站形式实现用户与系统之间的信息交互,主要功能有:首页、遥测数据、巡 测数据、卫星数据、预测预警、用户中心等功能。
立体监控展示在遥测数据、巡测数据和卫星数据三个网页中实现,各自的页面功能相似。立体监测网页通过与数据库相交互,向公众展示太湖立体监控系统采集的湖泊实时环境信息,主要加载控件和引用对象包括:数据库引擎、按钮、图表、复选框、下拉组合框、文本框、时间和WebGIS等。首先,该网页可以在电子地图上展示单个站点的实时气象、水文和水质监测数据,也可以通过电子地图二次开发网格化展示面状监测数据。其次,该网页还可以给不同权限用户提供历史数据检索、下载和简单统计分析服务。数据检索和下载是用户提供站点、指标和时间等信息,网站自动生成SQL语句向数据库发送命令,数据库收到命令而返回相应的数据集,通过图和表的形式向用户展示,也可以生成指定格式的文件供高级用户批量下载。简单的统计分析,网站提供了不超过1年的时间序列的统计分析,包括最大值、最小值、平均值、计数和求和等简单的统计功能。
预测预警网页通过与数据库相交互,向公众展示太湖蓝藻灾害预测预警信息,主要加载控件和引用对象包括:数据库引擎、时间、按钮、iWebOffice、文本框、下拉列表框、复选框、Frame、WebGIS和Flash等。该网页具有:1)将以网页表的形式展示三维数值模型涉及的计算网格、时间步长、预报期、初始值和边界条件等信息,也列出所有模型参数名称、功能、取值范围和当前值等信息。同时,用户可以通过内置iWebOffice插件对上述模型设置进行修改;2)绘制等值线图,该网页可以调取模型模拟数据,通过WebGIS绘制等值线图,同时在Flash控件以时间顺序播放这些等值线图,形成未来3d太湖模拟环境指标的时空变化动画;3)该网页能够按照预先指定的格式自动制作太湖水污染及蓝藻监测预警半周报。除在网页展示外,该半周报还可以通过超链接下载,也可以自动在指定时间向指定的电子邮箱地址发送。
用户中心网页主要实现对不同级别用户管理和用户自主发布新闻等功能,主要加载控件和引用对象包括:数据库引擎、按钮、Flash、图片和表等。首先该网页能够为所有用户设置不同的权限,包括系统管理员、高级用户、中级用户和初级用户。系统管理员拥有对网站的所有权限,可以制定所有规则。高级用户可以浏览网站所有页面、访问数据库,批量下载监测数据,自主发布新闻和操控三维数值模型。中级用户可以浏览网站所有页面、访问数据库和批量下载监测数据。初级用户就是一般访客,仅可以浏览网站向所有普通公众展示的蓝藻灾害信息。
实施例2
本实施例以太湖为例,对本发明的系统作进一步描述。
本发明所述面向湖泊蓝藻灾害的立体监控及数据挖掘系统包括监测子系统和数据处理子系统,所述监测子系统用于采集待监测数据;包括利用遥感进行监测的遥感监测子系统、利用自动监测站进行监测的自动监测子系统和人工采集待监测数据的人工巡测子系统;
所述遥感监测子系统包括卫星数据接收天线、计算机和供电系统,所述供电系统用于为耗电装置供电,卫星数据接收天线接收卫星数据,并向计算机传输,通过计算机进行遥感反演处理后,将处理后的数据通过互联网传输至数据处理子系统;
所述自动监测子系统为多个自动监测站通过无线网络连接形成的监测网络,单个自动监测站由水面支撑系统、供电系统、安全警示系统和数据采集系统组成;所述水面支撑系统为自动监测站硬件装置的承重结构;所述供电系统用于为自动监测站的耗电装置供电;所述安全警示系统用于安全警示,防止自动监测站遭受意外破坏;所述数据采集系统用于采集包括从气象仪器、水文仪器、水质仪器和视频仪器中获取的待监测数据,并将采集的数据通过无线网络传输到数据处理子系统;
所述人工巡测子系统为人工采集数据,并将数据通过互联网传输至数据处理子系统;
所述数据处理子系统用于接收和处理监测子系统获取的数据;包括服务器、阵列机、计算机、计算工作站、硬件防火墙、路由器、网线和供电设备;服务器采用双机热备模式,两台服 务器和用于数据存储的阵列机通过三叉电缆实现心跳连接,所有计算机、服务器和计算工作站均通过网线与硬件防火墙相连,硬件防火墙通过连接路由器与外网连通;
数据处理子系统接收数据后,对接收的数据进行数据备份和数据处理,包括数据预处理、三维数值模拟和蓝藻灾害评估,获取湖泊蓝藻灾害的预测指标和风险评估,并通过公共平台发布。所述数据预处理、三维数值模拟和蓝藻灾害评估算法均可参照实施例1选用的算法。
如图2所示,本实施例中的自动监测站是指18个分布于太湖蓝藻水华重点发生区域的自动监测站组成的监测网,可以30min的时间分辨率连续实时记录这18个自动监测站所在水域水环境信息。
所述供电设备采用UPS系统不间断电源供电,所述的两台服务器居于硬件防火墙和路由器下的局域网中,居于同一局域网中的还有用于遥感数据接收和处理的计算机,以及用于三维数值模型运行的计算工作站,计算工作站与两台服务器处于同一局域网中,可以满足三维数值模型对数据库数据的自动快速读取,并可将计算结果回传到Oracle数据库中,所有设备均通过电缆与UPS系统相连接,UPS系统则与民用交流电相连接,UPS系统为数据中心提供稳定的电源供应。
通过编写遥感数据接收程序将同处于一个局域网内的计算机上的卫星数据向服务器传送,并按照规定的文件命名模式存储阵列机中。自动监测站的通信模块中包括GRPS模块和CR1000数据采集器。只要在服务器端安装与CR1000相匹配的lognet软件就可以实现将自动监测数据传输到服务器端,并按照规定的文件命名模式存储阵列机中。

Claims (15)

  1. 一种面向湖泊蓝藻灾害的立体监控及数据挖掘方法,其特征在于,包括如下步骤:
    (1)通过遥感监测、自动监测和人工巡测三种途径获取与待监测湖泊蓝藻灾害相关的监测指标数据,包括气象指标、水文指标、水质指标和视频影像,获取的数据通过互联网传输至数据中心;
    其中,所述遥感监测指通过卫星遥感实时监测;
    所述自动监测指由多个自动监测站通过无线网络连接形成监测网络,对待监测指标进行实时监测;
    所述人工巡测指通过人工方式进行指标检测;
    (2)数据中心对接收的数据进行数据备份和数据预处理,所述数据预处理包括如下步骤:
    根据设定的数据采集时间间隔对数据进行检查,如果数据有中断,则通过时间插值方法对数据进行插值处理;
    对遥感监测获取的卫星数据,通过空间插值填补云层覆盖区域的数据缺失;
    对数据进行异常判定和处理,对于判定的异常数据,采用异常数据的前一个数据代替该异常数据;
    数据经预处理后,连同数据中心接收的原始数据一同传输至数据库存储;
    (3)根据数据库中存储的数据源构建待监测湖泊的三维数值模型,具体为:
    构建待监测湖泊的水动力模型;
    在待监测湖泊的水动力模型上叠加物质迁移转化模型,两种模型的方程耦合计算;其中,所述物质迁移转化模型计算的标量包括光照、悬移质、藻类生长、营养盐循环和溶解氧;
    采用有限差分求解模型,获取数值模型模拟数据;
    (4)根据数值模型模拟数据,进行待监测湖泊蓝藻灾害风险评估;
    (5)在公共平台上展示湖泊蓝藻灾害预测预警信息。
  2. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,对数据库中存储的数据进行分类存储,具体如下:
    对于单点时间连续的数据,以单个数据表存储单个监测站的所有数据;
    对于二维数据直接存放在数据表中;
    对于三维数值模型生成的数据,以时间为节点,存放在数据表中;
    对于图像或视频数据,将图像或视频存储在阵列机中,在数据库中建立数据表记录图像或视频的路径,字段包括编号、时间和图像/视频路径,以索引的方式访问图像或视频。
  3. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,数据库面向多源异构数据集构建,选用SQL Server、Access或Oracle;优选Oracle。
  4. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,所述时间插值方法采用线性插值、样条函数插值或分段插值;
    空间插值方法采用邻近点法、克里格法或反距离加权法;
    异常值判定和处理的依据为趋势检验、专家经验或数值比对。
  5. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,所述时间插值方法采用线性插值,算法具体如下:
    对于数据集中a1和a2两个数据,依据时间顺序,中间缺少b1,b2,……,bn数据,那么:
    Figure PCTCN2017089012-appb-100001
    其中i=[1,n]。
  6. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,所述空间插值方法采用 反距离加权法,算法具体如下:
    假设空间点坐标(x0,y0)处指标值缺测C(x0,y0),设定一个搜索半径,使得在此半径范围内至少包含3个数据点;然后使用这些已知数据点数据求取未知点的指标值:
    Figure PCTCN2017089012-appb-100002
    式中,C(x1,y1),C(x2,y2),…,C(xn,yn)分别表示括号内坐标点处实测指标值;d1,d2,…,dn分别表示括号内坐标点到空间点坐标(x0,y0)对应的直线距离,n≥3。
  7. 根据权利要求1所述的方法,其特征在于,所述步骤(2)中,采用5倍方差法进行异常值判定,具体方法如下:将第m个实测数据am及其前后5个数据求平均值和方差:
    Figure PCTCN2017089012-appb-100003
    Figure PCTCN2017089012-appb-100004
    Figure PCTCN2017089012-appb-100005
    为判断区间,满足
    Figure PCTCN2017089012-appb-100006
    的数据则为常规数据,否则以第m-1个数据代替am
  8. 根据权利要求1所述的方法,其特征在于,所述步骤(3)中,水动力模型的控制方程如下所示:
    Figure PCTCN2017089012-appb-100007
    Figure PCTCN2017089012-appb-100008
    Figure PCTCN2017089012-appb-100009
    Figure PCTCN2017089012-appb-100010
    Figure PCTCN2017089012-appb-100011
    式中:u,v,w分别为x,y,σ三个方向上流速;H和t为水深和时间;g和ξ为重力加速度和水位;f为科里奥利力,P为压力,Bx,By和BT分别为x向动量,y向动量和温度方程的由坐标转换而引入的小项;T为温度;Sh为外部进入系统的热量;Cp为水体的热容量;Kh,Kv为热量在水平及垂直方向上的扩散系数;ρ为水体密度,ρ0为水体参考密度;Ah为水平涡粘性系数;Av为垂向涡粘系数,采用下式定义:
    Figure PCTCN2017089012-appb-100012
    Figure PCTCN2017089012-appb-100013
    式中:v0、m0、m1值分别为5.0×10-6m2/s,0.1和-1;l为普朗特长度;Ri为Richardson数,反映流体稳定性状况,其表达式为:
    Figure PCTCN2017089012-appb-100014
    在水气界面σ=1的风能输入和水土界面σ=0的摩阻力分别表达为:
    Figure PCTCN2017089012-appb-100015
    Figure PCTCN2017089012-appb-100016
    式中:ρa和ρs分别表示空气密度和表层水体密度;CWD是风拖拽系数;WS是水表以上10m高度处风速;CSD为湖底摩擦系数;
    采用分裂算子技术求解水动力方程(5)、(6)和(7),生成内外两种模态;然后利用有限差分离散内外两种模态,其中水平和时间差分格式为显式,垂直差分格式为隐式,并采用低通滤波器对水面位移在时间方向上进行平滑处理;最后采用追赶法求解超大型稀疏矩阵。
  9. 根据权利要求8所述的方法,其特征在于,在水动力模型上叠加的物质迁移转化模型的控制方程如下所示:
    γPAR(I,J,K)=γ0,PAR1,PARCHLA(I,J,K)+γ2,PARSED(I,J,K)      (15)
    Figure PCTCN2017089012-appb-100017
    Figure PCTCN2017089012-appb-100018
    式中:γPAR(I,J,K)、γ0,PAR分别表示总衰减系数和纯水衰减系数;γ1,PAR、γ2,PAR分别表示藻类比衰减系数和非藻类颗粒物比衰减系数;CHLA(I,J,K)为以叶绿素a浓度表示的浮游植物生物量;SED(I,J,K)非藻类颗粒物浓度;u,v,w为新时间步的三维流速;S为悬移质浓度;ws为悬浮物沉降速率;J0为水土界面通量项,包括侵蚀通量和沉降通量;Ci表示第i种物质浓度;SKi表示生化过程项,i=1,2,3,4,5分别表示叶绿素a、浮游动物、可利用磷、可利用氮、溶解氧及五日生化需氧量;
    基于水动力模型计算结果,利用有限差分法在矩形网格中离散上述方程:空间上离散采用迎风格式;水平和时间差分格式为显式,垂直差分格式为隐式;追赶法求解超大型稀疏矩阵;
    方程(17)涉及的生化项采用下列方法计算:
    藻类生化项:
    Figure PCTCN2017089012-appb-100019
    式中:μ为总生长率;KM为非牧食死亡率;KS为上浮率;ZP为浮游动物滤食率;CHLAi,j,k表示i,j,k网格的浮游植物生物量;此处的τa、τf、τp及下述公式涉及的τZP,τKB,τs,τd,τso均为温度影响因子,表达式分别为θa T-20
    Figure PCTCN2017089012-appb-100020
    θp T-20,θZP T-20,θKB T-20,θs T-20,θd T-20及θso T-20,其中θa,θp,θZP,θKB,θs,θd和θso为温度乘子;
    浮游动物生化项:
    Figure PCTCN2017089012-appb-100021
    式中:μzp为浮游动物生长率;KCHLA为浮游动物牧食半饱和参数;BFISH和FISH分别表示鱼滤水率和鱼生物量;ZOOP表示浮游动物生物量;
    DTP和DTN生化项:
    Figure PCTCN2017089012-appb-100022
    Figure PCTCN2017089012-appb-100023
    式中:FMRP、FMRN、FMDP和FMDN分别为藻类代谢物磷转化率、代谢物氮转化率、死亡残骸磷转化率和死亡残骸氮转化率;KD和KM分别为藻类代谢率和死亡率;RPJ和RNJ分别沉积物磷和氮静态释放率;RPD和RND分别沉积物磷和氮动态释放率;SEDF水土界面悬浮物通量;ZDP和ZDN分别藻类对磷和氮吸收率;KPS和KNS为DTP和DTN的沉降率;
    DO和BOD生化项:
    Figure PCTCN2017089012-appb-100024
    SK6=(τNP·ZHD+τR·ZHR)CHLA-τKB·KB·BOD      (23)
    式中:KOD为大气复氧率;DOSAT为饱和DO;Hs为特征波高;PP藻类光合产氧;ZHY为藻类呼吸耗氧;RO为沉积物耗氧率;ZHD和ZHR为藻类死亡和代谢物BOD产生率;KB为BOD降解速率。
  10. 根据权利要求1所述的方法,其特征在于,所述步骤(3)中,三维数值模型从数据库中读取某时刻点的初始化数据和边界条件数据,并依据上个模拟期内的实测数据,采用Monte Carlo法对模型参数组合进行优化;
    在完成所需数据输入后,模型自动激活开展本轮模拟运算,并将本轮计算结果回传给数据库;
    在结束上述过程后,模型转入休眠状态,等待下轮模拟运算。
  11. 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统,包括监测子系统和数据处理子系统,其特征在于,
    所述监测子系统用于采集待监测数据;包括利用遥感进行监测的遥感监测子系统、利用自动监测站进行监测的自动监测子系统和人工采集待监测数据的人工巡测子系统;
    所述遥感监测子系统包括卫星数据接收天线、计算机和供电系统,所述供电系统用于为耗电装置供电,卫星数据接收天线接收卫星数据,并向计算机传输,通过计算机进行遥感反演处理后,将处理后的数据通过互联网传输至数据处理子系统;
    所述自动监测子系统为多个自动监测站通过无线网络连接形成的监测网络,单个自动监测站由水面支撑系统、供电系统、安全警示系统和数据采集系统组成;所述水面支撑系统为自动监测站硬件装置的承重结构;所述供电系统用于为自动监测站的耗电装置供电;所述安全警示系统用于安全警示,防止自动监测站遭受意外破坏;所述数据采集系统用于采集包括从气象仪器、水文仪器、水质仪器和视频仪器中获取的待监测数据,并将采集的数据通过无线网络传输到数据处理子系统;
    所述人工巡测子系统为人工采集数据,并将数据通过互联网传输至数据处理子系统;
    所述数据处理子系统用于接收和处理监测子系统获取的数据;包括服务器、阵列机、计算机、计算工作站、硬件防火墙、路由器、网线和供电设备;服务器采用双机热备模式,两台服务器和用于数据存储的阵列机通过三叉电缆实现心跳连接,所有计算机、服务器和计算工作站均通过网线与硬件防火墙相连,硬件防火墙通过连接路由器与外网连通;
    数据处理子系统接收数据后,对接收的数据进行数据备份和数据处理,包括数据预处理、三维数值模拟和蓝藻灾害评估,获取湖泊蓝藻灾害的预测指标和风险评估,并通过公共平台发 布。
  12. 根据权利要求11所述的系统,其特征在于,所述供电设备为采用UPS系统的不间断电源供电设备。
  13. 根据权利要求11所述的系统,其特征在于,数据处理子系统中,两台服务器、用于遥感数据接收和处理的计算机、以及用于三维数值模型运行的计算工作站居于同一局域网中。
  14. 根据权利要求11所述的系统,其特征在于,所述数据处理子系统接收监测子系统发送的数据后,对数据进行数据备份和数据预处理,所述数据预处理包括如下步骤:
    根据设定的数据采集时间间隔对数据进行检查,如果数据有中断,则通过时间插值方法对数据进行插值处理;
    对遥感监测获取的卫星数据,通过空间插值填补云层覆盖区域的数据缺失;
    使用5倍方差法对数据进行异常判定和处理,对于判定的异常数据,采用异常数据的前一个数据代替该异常数据;
    数据经预处理后,连同数据中心接收的原始数据一同传输至数据库存储。
  15. 根据权利要求11所述的系统,其特征在于,两台服务器和阵列机中安装有Oracle数据库软件;Oracle数据库存储遥感监测、自动监测、人工巡测和用于三维数值模拟的数据;并对数据进行分类存储,具体如下:
    对于单点时间连续的数据,以单个数据表存储单个监测站的所有数据;
    对于二维数据直接存放在数据表中;
    对于三维数值模型生成的数据,以时间为节点,存放在数据表中;
    对于图像或视频数据,将图像或视频存储在阵列机中,在数据库中建立数据表记录图像或视频的路径,字段包括编号、时间和图像/视频路径,以索引的方式访问图像或视频。
PCT/CN2017/089012 2017-06-19 2017-06-19 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法 WO2018214190A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/329,943 US11402362B2 (en) 2017-06-19 2017-06-19 Stereoscopic monitoring and data mining system and method for harmful lake cyanobacteria bloom
PCT/CN2017/089012 WO2018214190A1 (zh) 2017-06-19 2017-06-19 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/089012 WO2018214190A1 (zh) 2017-06-19 2017-06-19 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法

Publications (1)

Publication Number Publication Date
WO2018214190A1 true WO2018214190A1 (zh) 2018-11-29

Family

ID=64396109

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/089012 WO2018214190A1 (zh) 2017-06-19 2017-06-19 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法

Country Status (2)

Country Link
US (1) US11402362B2 (zh)
WO (1) WO2018214190A1 (zh)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111159629A (zh) * 2019-12-31 2020-05-15 西安建筑科技大学 一种分层湖泊水库水体完全混合的碳排放量的计算方法
CN111221806A (zh) * 2020-01-20 2020-06-02 南京大学(溧水)生态环境研究院 一种物质循环过程模拟数据库的构建方法
CN111274938A (zh) * 2020-01-19 2020-06-12 四川省自然资源科学研究院 一种面向Web的高分遥感河流水质动态监测方法与系统
CN112557616A (zh) * 2020-12-10 2021-03-26 中国科学院水生生物研究所 一种通江湖泊商业采砂活动影响水生态健康的监测方法
CN112763429A (zh) * 2020-12-30 2021-05-07 济南工程职业技术学院 一种蓝藻水华的立体监测方法
CN112763426A (zh) * 2020-12-23 2021-05-07 宁德卫星大数据科技有限公司 一种循环优化的高光谱大数据水质全天候动态监测方法
CN112989692A (zh) * 2021-02-10 2021-06-18 中国科学院南京地理与湖泊研究所 基于遥感数据的湖泊富营养化反演方法
CN113158762A (zh) * 2021-02-23 2021-07-23 中国科学院南京地理与湖泊研究所 大型湖库滨岸带蓝藻水华堆积风险评估方法
CN113516373A (zh) * 2021-06-18 2021-10-19 中国科学院南京地理与湖泊研究所 蓝藻水华防控集成技术体系
CN113673737A (zh) * 2020-05-14 2021-11-19 中国科学院南京地理与湖泊研究所 基于卫星遥感影像的藻型湖泊水体溶解二氧化碳估算方法
CN113687609A (zh) * 2021-07-21 2021-11-23 浙江微科机电有限公司 应用于异常环境下的物联网智能监测系统及监测方法
CN113763272A (zh) * 2021-09-03 2021-12-07 中国科学院南京地理与湖泊研究所 一种富营养化湖泊光合有效辐射衰减系数遥感反演方法
CN113947726A (zh) * 2021-10-29 2022-01-18 江苏天汇空间信息研究院有限公司 基于互联网的生态河湖区域监管系统
CN114720582A (zh) * 2021-11-26 2022-07-08 韩山师范学院 一种不同陈化年份老香黄的综合评价方法
CN115032355A (zh) * 2022-06-07 2022-09-09 中国地质调查局南京地质调查中心(华东地质科技创新中心) 一种多深度地下水水质参数自动化监测系统
CN115578646A (zh) * 2022-09-16 2023-01-06 二十一世纪空间技术应用股份有限公司 一种蓝藻水华监测方法及装置
CN115855159A (zh) * 2022-12-16 2023-03-28 水利部交通运输部国家能源局南京水利科学研究院 一种柔性漂浮物漂浮特性监测装置及监测方法
CN115906505A (zh) * 2022-12-12 2023-04-04 武汉大学 基于gnss信号的水面水华密度遥感探测方法及系统
CN115907471A (zh) * 2022-11-22 2023-04-04 安徽农业大学 基于VineCopula函数的湖泊蓝藻风险诊断方法
CN115936496A (zh) * 2022-11-29 2023-04-07 中国环境科学研究院 水质预测模型数据治理标准化方法
CN116859815A (zh) * 2023-09-04 2023-10-10 深圳市水务技术服务有限公司 一种用于水库泵站的实时监控预警系统及方法
CN116952906A (zh) * 2023-09-20 2023-10-27 南京航天宏图信息技术有限公司 水体健康状态评估方法、装置、电子设备及存储介质
CN117113735A (zh) * 2023-10-25 2023-11-24 湖南大学 基于多层次生态模型的藻华智能预警方法与系统
CN117434034A (zh) * 2023-10-24 2024-01-23 上海普适导航科技股份有限公司 一种基于光谱库的水质多源遥感数据快速反演方法
CN113763272B (zh) * 2021-09-03 2024-06-11 中国科学院南京地理与湖泊研究所 一种富营养化湖泊光合有效辐射衰减系数遥感反演方法

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114088907A (zh) * 2021-11-05 2022-02-25 常州市环境科学研究院 一种水、陆、空一体化的浅水湖泊蓝藻水华监测及预警系统
CN114018338B (zh) * 2021-11-17 2023-08-22 天津市水利科学研究院 基于光谱指数模型的水体识别系统
CN114399415B (zh) * 2021-12-23 2023-04-25 广东贝源检测技术股份有限公司 一种生物多样性的调查评估系统
CN114564883B (zh) * 2022-02-11 2023-05-19 水利部交通运输部国家能源局南京水利科学研究院 融合机理与ML的湖泊叶绿素a短期集合预报方法及系统
CN114538705A (zh) * 2022-02-21 2022-05-27 中国科学院地理科学与资源研究所 一种湖泊长时序连续水域变化重建方法
CN114814136B (zh) * 2022-04-14 2024-05-28 水利部交通运输部国家能源局南京水利科学研究院 一种rue界定城市河网水华程度方法及其应用
CN114880524A (zh) * 2022-05-05 2022-08-09 深圳艾灵网络有限公司 基于数据湖的数据通信方法、电子设备及存储介质
CN114998724B (zh) * 2022-05-13 2024-04-23 安徽农业大学 一种抗云层干扰的湖泊蓝藻水华遥感监测方法
CN114994274A (zh) * 2022-05-30 2022-09-02 佛山市美的清湖净水设备有限公司 软水设备的原水硬度确定方法及装置
CN114897422B (zh) * 2022-06-01 2023-07-11 中国长江三峡集团有限公司 蓝藻水华损失评估方法、装置、电子设备及存储介质
CN114821977B (zh) * 2022-06-27 2022-09-02 环球数科集团有限公司 一种基于卫星遥感监测的生态监测灾害预警系统
CN114881544B (zh) * 2022-07-07 2022-09-30 中国长江三峡集团有限公司 一种水库流量调控方法、装置、电子设备及存储介质
CN115330132B (zh) * 2022-07-20 2023-04-07 中交上海航道局有限公司 一种用于突发污染事故中宽浅型河流的水质分布逆时反演的方法
CN115423157B (zh) * 2022-08-15 2023-04-28 中国水利水电科学研究院 一种山洪灾害风险动态预警方法及装置
CN115062874B (zh) * 2022-08-16 2022-11-18 成都禀证科技有限责任公司 一种水体污染物监测预测分析方法及系统
CN115146484A (zh) * 2022-08-31 2022-10-04 天津市环鉴环境检测有限公司 一种用于检测环境参数的环保监控系统及监控方法
CN115951025B (zh) * 2022-09-07 2023-07-14 北京智科远达数据技术有限公司 一种水体总磷浓度的反演方法
CN115901640B (zh) * 2022-10-19 2024-03-15 山东大学 融合光谱成像与时空分布的不良地质超前预报方法及系统
CN115953687B (zh) * 2023-01-18 2023-11-10 生态环境部卫星环境应用中心 基于遥感技术的小微水体受损等级划分方法和装置
CN116186144B (zh) * 2023-03-14 2023-10-27 中国自然资源航空物探遥感中心 矿山遥感监测数据的格式化自动处理方法及系统
CN117890546A (zh) * 2023-04-11 2024-04-16 江苏省秦淮河水利工程管理处 一种湖泊生态风险预警和干预方法及系统
CN116482317B (zh) * 2023-04-26 2023-10-27 大连理工大学 湖泊水体营养状态实时监测方法、系统、设备及介质
CN116577832A (zh) * 2023-05-12 2023-08-11 中国科学院东北地理与农业生态研究所 基于无人机载微波辐射观测系统的高分辨率湖冰探测方法
CN116467565B (zh) * 2023-06-20 2023-09-22 国家海洋局北海预报中心((国家海洋局青岛海洋预报台)(国家海洋局青岛海洋环境监测中心站)) 一种浒苔绿潮斑块最优搜寻区域预报方法
CN116819024B (zh) * 2023-06-28 2024-04-19 同济大学 一种湖库富营养化生态动力学模型的模拟优化方法及装置
CN116502880B (zh) * 2023-06-29 2023-10-20 长江三峡集团实业发展(北京)有限公司 一种考虑湖泊水质动态响应的水库生态调度方法及装置
CN117035164B (zh) * 2023-07-10 2024-03-12 江苏省地质调查研究院 一种生态灾害物监测方法及系统
CN116879192B (zh) * 2023-09-07 2023-12-12 航天宏图信息技术股份有限公司 基于卫星遥感数据的水华预测方法、装置、设备及介质
CN117030966B (zh) * 2023-10-09 2023-12-12 北京林业大学 基于大数据的生态环境智能监测方法及系统
CN117114428B (zh) * 2023-10-25 2024-01-30 国网山西省电力公司电力科学研究院 一种电力设备气象灾害分析预警方法
CN117391374B (zh) * 2023-10-30 2024-03-08 江苏国贸酝领智能科技股份有限公司 一种基于物联网技术的智慧水务监管方法及系统
CN117390466B (zh) * 2023-12-13 2024-02-06 江西省水利科学院(江西省大坝安全管理中心、江西省水资源管理中心) 一种基于相似性度量的湖泊稳态判别方法
CN117745754A (zh) * 2023-12-18 2024-03-22 中国科学院空天信息创新研究院 一种蓝藻水华自动化监测方法和系统
CN117633721B (zh) * 2024-01-25 2024-04-09 水利部交通运输部国家能源局南京水利科学研究院 机理模型与数据联合驱动的城市河网透明度预测方法
CN117853938B (zh) * 2024-03-08 2024-05-10 鲸服科技有限公司 基于图像识别的生态监测系统及方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103063202A (zh) * 2012-12-30 2013-04-24 同济大学 基于遥感影像的蓝藻生物量时空变化监测与可视化方法
CN103678910A (zh) * 2013-12-12 2014-03-26 河海大学 河道型水库支流库湾水华的云体系架构预警系统及方法
WO2016090883A1 (zh) * 2014-12-12 2016-06-16 中国矿业大学 一种基于多源信息融合的采场顶板离层水害超前预报方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6536272B1 (en) * 1999-08-06 2003-03-25 University Of Miami Water monitoring, data collection, and transmission module
JP2003533690A (ja) * 2000-05-16 2003-11-11 ジークル・リミテッド 天然水の測光分析
US20120058542A1 (en) * 2009-03-11 2012-03-08 Live Fuels ,Inc. Systems and methods for regulating algal biomass
CN103604423B (zh) * 2013-11-22 2015-07-01 中国科学院南京地理与湖泊研究所 一种浅水湖泊遥感野外自动监测系统及监测方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103063202A (zh) * 2012-12-30 2013-04-24 同济大学 基于遥感影像的蓝藻生物量时空变化监测与可视化方法
CN103678910A (zh) * 2013-12-12 2014-03-26 河海大学 河道型水库支流库湾水华的云体系架构预警系统及方法
WO2016090883A1 (zh) * 2014-12-12 2016-06-16 中国矿业大学 一种基于多源信息融合的采场顶板离层水害超前预报方法

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
KONG ET AL: "The theory and practice of prevention, forecast and warning on cyanobacteria bloom in Lake Taihu", JOURNAL OF LAKE SCIENCES, vol. 21, no. 3, 31 December 2009 (2009-12-31), pages 314 - 328, ISSN: 1003-5427 *
QIN ET AL: "The Application of A High-frequency, Wireless and Automatic Observation Technique in Algal Bloom Monitoring and Forecasting in Eutrophic Lake", E- SCIENCE, no. 3, 30 September 2010 (2010-09-30), pages 61 - 70, ISSN: 1674-9480 *
SHEN ET AL: "Research and Application of the Monitoring and Forecasting for Cyanobacterial Blooms and the Key Technologies for the Protection of Drinking Water Source: Introduction and Implication", SCIENCE & TECHNOLOGY FOR DEVELOPMENT, vol. 12, no. 1, 31 December 2016 (2016-12-31), ISSN: 1672-996X *

Cited By (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111159629A (zh) * 2019-12-31 2020-05-15 西安建筑科技大学 一种分层湖泊水库水体完全混合的碳排放量的计算方法
CN111159629B (zh) * 2019-12-31 2023-08-15 西安建筑科技大学 一种分层湖泊水库水体完全混合的碳排放量的计算方法
CN111274938A (zh) * 2020-01-19 2020-06-12 四川省自然资源科学研究院 一种面向Web的高分遥感河流水质动态监测方法与系统
CN111221806A (zh) * 2020-01-20 2020-06-02 南京大学(溧水)生态环境研究院 一种物质循环过程模拟数据库的构建方法
CN111221806B (zh) * 2020-01-20 2024-04-16 南京大学(溧水)生态环境研究院 一种物质循环过程模拟数据库的构建方法
CN113673737B (zh) * 2020-05-14 2023-07-28 中国科学院南京地理与湖泊研究所 基于卫星遥感影像的藻型湖泊水体溶解二氧化碳估算方法
CN113673737A (zh) * 2020-05-14 2021-11-19 中国科学院南京地理与湖泊研究所 基于卫星遥感影像的藻型湖泊水体溶解二氧化碳估算方法
CN112557616A (zh) * 2020-12-10 2021-03-26 中国科学院水生生物研究所 一种通江湖泊商业采砂活动影响水生态健康的监测方法
CN112763426A (zh) * 2020-12-23 2021-05-07 宁德卫星大数据科技有限公司 一种循环优化的高光谱大数据水质全天候动态监测方法
CN112763429A (zh) * 2020-12-30 2021-05-07 济南工程职业技术学院 一种蓝藻水华的立体监测方法
CN112989692A (zh) * 2021-02-10 2021-06-18 中国科学院南京地理与湖泊研究所 基于遥感数据的湖泊富营养化反演方法
CN113158762B (zh) * 2021-02-23 2024-05-28 中国科学院南京地理与湖泊研究所 大型湖库滨岸带蓝藻水华堆积风险评估方法
CN113158762A (zh) * 2021-02-23 2021-07-23 中国科学院南京地理与湖泊研究所 大型湖库滨岸带蓝藻水华堆积风险评估方法
CN113516373A (zh) * 2021-06-18 2021-10-19 中国科学院南京地理与湖泊研究所 蓝藻水华防控集成技术体系
CN113687609A (zh) * 2021-07-21 2021-11-23 浙江微科机电有限公司 应用于异常环境下的物联网智能监测系统及监测方法
CN113763272A (zh) * 2021-09-03 2021-12-07 中国科学院南京地理与湖泊研究所 一种富营养化湖泊光合有效辐射衰减系数遥感反演方法
CN113763272B (zh) * 2021-09-03 2024-06-11 中国科学院南京地理与湖泊研究所 一种富营养化湖泊光合有效辐射衰减系数遥感反演方法
CN113947726A (zh) * 2021-10-29 2022-01-18 江苏天汇空间信息研究院有限公司 基于互联网的生态河湖区域监管系统
CN113947726B (zh) * 2021-10-29 2022-07-01 江苏天汇空间信息研究院有限公司 基于互联网的生态河湖区域监管系统
CN114720582A (zh) * 2021-11-26 2022-07-08 韩山师范学院 一种不同陈化年份老香黄的综合评价方法
CN114720582B (zh) * 2021-11-26 2023-10-20 韩山师范学院 一种不同陈化年份老香黄的综合评价方法
CN115032355A (zh) * 2022-06-07 2022-09-09 中国地质调查局南京地质调查中心(华东地质科技创新中心) 一种多深度地下水水质参数自动化监测系统
CN115032355B (zh) * 2022-06-07 2023-07-21 中国地质调查局南京地质调查中心(华东地质科技创新中心) 一种多深度地下水水质参数自动化监测系统
CN115578646A (zh) * 2022-09-16 2023-01-06 二十一世纪空间技术应用股份有限公司 一种蓝藻水华监测方法及装置
CN115907471A (zh) * 2022-11-22 2023-04-04 安徽农业大学 基于VineCopula函数的湖泊蓝藻风险诊断方法
CN115936496A (zh) * 2022-11-29 2023-04-07 中国环境科学研究院 水质预测模型数据治理标准化方法
CN115936496B (zh) * 2022-11-29 2023-09-19 中国环境科学研究院 水质预测模型数据治理标准化方法
CN115906505A (zh) * 2022-12-12 2023-04-04 武汉大学 基于gnss信号的水面水华密度遥感探测方法及系统
CN115906505B (zh) * 2022-12-12 2024-05-24 武汉大学 基于gnss信号的水面水华密度遥感探测方法及系统
CN115855159B (zh) * 2022-12-16 2023-11-03 水利部交通运输部国家能源局南京水利科学研究院 一种柔性漂浮物漂浮特性监测装置及监测方法
CN115855159A (zh) * 2022-12-16 2023-03-28 水利部交通运输部国家能源局南京水利科学研究院 一种柔性漂浮物漂浮特性监测装置及监测方法
CN116859815B (zh) * 2023-09-04 2023-11-03 深圳市水务技术服务有限公司 一种用于水库泵站的实时监控预警系统及方法
CN116859815A (zh) * 2023-09-04 2023-10-10 深圳市水务技术服务有限公司 一种用于水库泵站的实时监控预警系统及方法
CN116952906B (zh) * 2023-09-20 2024-01-12 南京航天宏图信息技术有限公司 水体健康状态评估方法、装置、电子设备及存储介质
CN116952906A (zh) * 2023-09-20 2023-10-27 南京航天宏图信息技术有限公司 水体健康状态评估方法、装置、电子设备及存储介质
CN117434034A (zh) * 2023-10-24 2024-01-23 上海普适导航科技股份有限公司 一种基于光谱库的水质多源遥感数据快速反演方法
CN117113735A (zh) * 2023-10-25 2023-11-24 湖南大学 基于多层次生态模型的藻华智能预警方法与系统
CN117113735B (zh) * 2023-10-25 2024-01-02 湖南大学 基于多层次生态模型的藻华智能预警方法与系统

Also Published As

Publication number Publication date
US11402362B2 (en) 2022-08-02
US20210293770A1 (en) 2021-09-23

Similar Documents

Publication Publication Date Title
WO2018214190A1 (zh) 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法
CN107340365B (zh) 一种面向湖泊蓝藻灾害的立体监控及数据挖掘系统和方法
Zhu et al. Spatiotemporal analysis of hydrological variations and their impacts on vegetation in semiarid areas from multiple satellite data
Zeng et al. Population spatialization in China based on night-time imagery and land use data
Li et al. Remote sensing of human beings–a perspective from nighttime light
Yang et al. Spatial improvement of human population distribution based on multi-sensor remote-sensing data: An input for exposure assessment
Chen et al. A high-resolution monitoring approach of canopy urban heat island using a random forest model and multi-platform observations
CN107271627A (zh) 一种基于物联网和大数据技术的可视化水质信息监测系统
Borgogno Mondino et al. Site selection of large ground-mounted photovoltaic plants: a GIS decision support system and an application to Italy
KR20160072432A (ko) 해양환경 관측자료 제공시스템
CN104239959A (zh) 地质灾害预测系统
CN112217897A (zh) 一种基于物联网平台的智慧水资源管理系统
Chang et al. A case study for the application of an operational two-dimensional real-time flooding forecasting system and smart water level gauges on roads in Tainan City, Taiwan
Bahamon et al. The new pelagic Operational Observatory of the Catalan Sea (OOCS) for the multisensor coordinated measurement of atmospheric and oceanographic conditions
CN102307399A (zh) 基于Phidgets传感控制的海洋环境质量监测系统
Smith et al. Decadal change in sediment community oxygen consumption in the abyssal northeast Pacific
Rey‐Sanchez et al. Detecting hot spots of methane flux using footprint‐weighted flux maps
CN113887974A (zh) 一种基于多源时空数据集成的城市资源环境承载力综合评价方法
Bai et al. Single-vessel plume dispersion simulation: Method and a case study using CALPUFF in the Yantian port area, Shenzhen (China)
Ge et al. Estimation of land surface heat fluxes based on Landsat 7 ETM+ data and field measurements over the northern Tibetan Plateau
Mamassis et al. OpenHi. net: A synergistically built, national-scale infrastructure for monitoring the surface waters of Greece
Liu et al. Conflict or Coordination? The Spatiotemporal Relationship Between Humans and Nature on the Qinghai‐Tibet Plateau
Ou et al. Territorial pattern evolution and its comprehensive carrying capacity evaluation in the coastal area of Beibu Gulf, China
Yang et al. Spatiotemporal characteristics of human activity on coastal landscape of Laizhou Bay
Zhao et al. Soil salinity inversion model based on BPNN optimization algorithm for UAV multispectral remote sensing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17911090

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17911090

Country of ref document: EP

Kind code of ref document: A1