CN113723704B - Water quality rapid prediction method based on continuous and graded mixed data - Google Patents
Water quality rapid prediction method based on continuous and graded mixed data Download PDFInfo
- Publication number
- CN113723704B CN113723704B CN202111044820.7A CN202111044820A CN113723704B CN 113723704 B CN113723704 B CN 113723704B CN 202111044820 A CN202111044820 A CN 202111044820A CN 113723704 B CN113723704 B CN 113723704B
- Authority
- CN
- China
- Prior art keywords
- data
- water quality
- sub
- parameters
- basin
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 141
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000013135 deep learning Methods 0.000 claims abstract description 36
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 238000009826 distribution Methods 0.000 claims abstract description 25
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 239000002352 surface water Substances 0.000 claims abstract description 6
- 238000011161 development Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 44
- 239000013598 vector Substances 0.000 claims description 27
- 239000010865 sewage Substances 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 22
- 238000001704 evaporation Methods 0.000 claims description 18
- 230000008020 evaporation Effects 0.000 claims description 18
- 238000005192 partition Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 14
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 11
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 10
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 9
- 229910052760 oxygen Inorganic materials 0.000 claims description 9
- 239000001301 oxygen Substances 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 5
- XKMRRTOUMJRJIA-UHFFFAOYSA-N ammonia nh3 Chemical compound N.N XKMRRTOUMJRJIA-UHFFFAOYSA-N 0.000 claims description 5
- WPBNNNQJVZRUHP-UHFFFAOYSA-L manganese(2+);methyl n-[[2-(methoxycarbonylcarbamothioylamino)phenyl]carbamothioyl]carbamate;n-[2-(sulfidocarbothioylamino)ethyl]carbamodithioate Chemical compound [Mn+2].[S-]C(=S)NCCNC([S-])=S.COC(=O)NC(=S)NC1=CC=CC=C1NC(=S)NC(=O)OC WPBNNNQJVZRUHP-UHFFFAOYSA-L 0.000 claims description 5
- 229910052757 nitrogen Inorganic materials 0.000 claims description 5
- 229910052698 phosphorus Inorganic materials 0.000 claims description 5
- 239000011574 phosphorus Substances 0.000 claims description 5
- 239000000126 substance Substances 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 3
- 244000157795 Cordia myxa Species 0.000 claims description 2
- 235000004257 Cordia myxa Nutrition 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims 1
- 238000004088 simulation Methods 0.000 abstract description 7
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000002354 daily effect Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 235000013305 food Nutrition 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000000306 recurrent effect Effects 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 239000000356 contaminant Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 244000025254 Cannabis sativa Species 0.000 description 1
- 240000004859 Gamochaeta purpurea Species 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 239000010841 municipal wastewater Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000004065 wastewater treatment Methods 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a water quality rapid prediction method based on continuous and graded mixed data, which predicts water quality changes of different areas in a river basin by using classification grades of parameters such as river basin land utilization area, GDP, population and the like, constructs a deep learning neural network, and rapidly calculates to obtain surface water quality changes and spatial distribution of the river basin under the condition of different grades of social and economic development as a basis for river basin water environment treatment. Compared with the prior art, the invention has the beneficial effects that: the water quality prediction method provided by the invention avoids the problem of social and economic statistical data loss when water quality prediction is carried out by adopting social and economic data, can carry out prediction by only using social and economic classification grade data without accurate social and economic data, and has stronger practicability. And the water quality is predicted through the deep learning neural network, and compared with the traditional numerical simulation method, the calculation speed is higher, and the application range is wider.
Description
Technical Field
The invention relates to the technical field of water quality prediction, in particular to a water quality rapid prediction method based on continuous and graded mixed data.
Background
With the increase of social and economic development and human pollution discharge activities, water pollution becomes an important factor for restricting the sustainable development of the society in China and the pursuit of high-quality life of people. However, with the continuous enhancement of social and economic relations among regions and the influence of climate change, the water environment evolution of China shows the complicated change of cross-region and multi-factor coupling influence, the water quality prediction and water environment treatment of a single water sample or river reach can not meet the requirements gradually, and the comprehensive water quality prediction with a large space range and multiple influence factors is urgently needed.
The traditional water quality prediction method mainly comprises a numerical simulation method and a mathematical statistics method. However, due to the complexity of the watershed hydrodynamic process and the contaminant water chemistry process, the transport conversion mechanism of water contaminants is not completely understood, resulting in a poor accuracy of numerical simulation. In addition, the numerical simulation method has a large amount of calculation, and has a certain difficulty in wide-range application. In the aspect of a mathematical statistical method, although a calculation quantity numerical simulation method is low, the requirements on data time sequence, continuity and normal distribution are strict, modeling is relatively difficult, and applicability is not strong.
Meanwhile, the existing water quality prediction method based on deep learning is based on a recurrent neural network, such as a long-term and short-term memory model (LSTM), the future water quality change is generally predicted according to historical water quality data, only the change trend of a single water quality index or the influence between water quality indexes is usually considered, and the influence of external factors such as social economy and the like on the water quality change is not considered, so that the support effect of the prediction method in water environment treatment decision and economic and social development planning of a decision maker is insufficient, and the practicability is low.
In addition, due to the complexity of the human socioeconomic system, socioeconomic indexes related to water quality are many, including GDP, population, food yield, cultivated land area, sewage discharge amount, sewage treatment degree, etc., and data loss or incomplete statistics in a certain period or a certain region are often encountered in the data statistics of the regional socioeconomic system. The invention adopts the socioeconomic classification grade data to predict the water quality, avoids the problem of social and economic statistical data loss, can predict the water quality only by using the grade data of the socioeconomic index without needing an accurate socioeconomic index value, and has stronger practicability.
Disclosure of Invention
In order to solve the defects of singleness and poor applicability of a water quality prediction method in the background art, and low practicability of prediction by only adopting a single index, the invention provides a water quality rapid prediction method based on continuous and graded mixed data. The method integrates social and economic classification grade data and has the advantages of high calculation speed and high operation efficiency.
In order to achieve the purpose, the technical scheme of the watershed water quality rapid prediction method based on classification data deep learning comprises the following steps:
a water quality rapid prediction method based on continuous and graded mixed data comprises the following steps:
s1, determining a drainage basin range and an internal partition thereof, wherein the internal partition comprises a sub-drainage basin partition and an administrative partition;
s2, determining meteorological parameters of the drainage basin in a prediction time period, and converting the meteorological parameters into sub-drainage basin meteorological data; the meteorological parameters comprise rainfall, evaporation and air temperature data observed at meteorological points at time i;
s3, determining social and economic parameters of the sub-watershed in a prediction time period, and generating a social and economic parameter set of the sub-watershed;
s4, determining water quality parameters of the sub-watershed;
s5, carrying out grade classification on data in the social and economic parameter set of the sub-basin;
s6, carrying out thermal coding on the social and economic parameters of the grade classification to form a thermal coding matrix of the social and economic parameters;
s7, carrying out standardization treatment on the meteorological parameters and the water quality parameters;
s8, dividing a training set and a testing set in a thermal coding matrix of social and economic parameters, standardized sub-basin meteorological data and water quality parameters;
s9, constructing a full-connection deep learning neural network, and defining a loss function and an iterative optimization algorithm;
s10, inputting the training set into a deep learning neural network to obtain a trained deep learning neural network, and then inputting the test set into the trained deep learning neural network to obtain a predicted water quality parameter; comparing the predicted water quality parameters with the actually measured water quality parameters, adjusting model parameters, and finally storing the deep learning neural network and parameter data thereof which meet the precision requirement;
s11, selecting meteorological and social economic data of a certain time period of the drainage basin, inputting the meteorological and social economic data into the deep learning neural network stored in the S10, and predicting to obtain water quality parameters of all the sub-drainage basins in the time period.
Further, step S2 is more specifically:
s21, acquiring rainfall, evaporation and air temperature data in a prediction period by a meteorological site;
s22, manufacturing a meteorological site space distribution map according to longitude and latitude coordinates of meteorological sites, and importing rainfall, evaporation and air temperature data of each site and each time period into an attribute table of the site distribution map;
s23, with the whole drainage basin as a boundary, performing spatial interpolation on meteorological data of each station by adopting a Krigin interpolation method to generate grid data of rainfall, evaporation and air temperature of the drainage basin in each period, and storing the grid data into a geospatial database;
and S24, calculating the average value of rainfall, evaporation and air temperature data of all grids in each sub-basin range by taking the sub-basins as ranges to obtain meteorological parameters of the sub-basins.
Further, the socioeconomic parameters include land utilization area, population living, GDP, population average GDP, population density, grain yield, and municipal sewage treatment rate.
Further, step S3 is more specifically:
s31, obtaining social and economic parameters of land utilization area, regular living population, GDP, average population GDP, population density, grain yield and urban sewage treatment rate in a prediction period;
s32, converting the social and economic parameters in the S31 into related parameters of the sub-watershed;
and S33, generating a socioeconomic parameter set of the sub-watershed.
Further, the water quality parameters comprise concentration values of dissolved oxygen, COD (chemical oxygen demand) manganese, ammonia nitrogen, total phosphorus and total nitrogen of the surface water; step S4 is more specifically:
s41, determining longitude and latitude coordinates of riverway water quality monitoring stations in all the sub-areas to manufacture a space distribution map of the water quality monitoring stations, and then importing the water quality index data of each riverway water quality monitoring station and each time period into an attribute table of the station distribution map;
s42, calculating water quality distribution in the river channel of the sub-basin;
s43, taking the sub-watersheds as ranges, and calculating the average value of the river water quality in each sub-watersheds range to serve as the water quality parameter of the sub-watersheds.
Further, step S42 is more specifically;
s421, collecting river network data of rivers with more than five levels in the whole river basin;
and S422, with the river network of the river basin as a boundary, performing spatial interpolation on the water quality data of each time interval of the station by adopting an inverse distance weighting method to generate water quality grid data of each time interval along the river, and storing the water quality grid data into a geospatial database.
Further, the land utilization area comprises utilization areas of cultivated land, forest land, grassland, water area, city and residential land and unused land; step S5 is more specifically: calculating grade values of the socioeconomic variables to form socioeconomic grade variables, wherein the grade values are the areas of cultivated land, forest land, grassland, water area, city and residential land, unused land, and the grade thresholds of socioeconomic parameters of the permanent population, GDP, grain yield, per capita GDP, population density and urban sewage treatment rate.
Further, step S6 is more specifically:
s61, converting the socioeconomic grade variable of each sub-basin and each time interval into 0-1 row vector; the length of the row vector is the total number N of the socioeconomic variables, namely the row vector is 1 row and N columns; in the row vector, the value of the nth column corresponding to the level value n of the social and economic level variable is 1; the values of the other elements in the row vector are 0;
s62, constructing a thermal coding 0-1 matrix of the socioeconomic grade variable.
Further, step S8 is more specifically: combining the thermal coding 0-1 matrixes of the sub-basin meteorological data and the social and economic data into a digital matrix which is used as an input digital matrix of the deep learning neural network; converting the sub-basin water quality parameters into a digital matrix as an output label matrix of the deep learning neural network; then 70% of the rows in the input digit matrix and the output label matrix are used as a training set, and 30% of the rows are used as a testing set.
Further, the parameters of the deep learning neural network comprise the number k of hidden layers of the network, the number n of nodes of each hidden layer, an activation function, a loss function, an accuracy calculation function, a learning rate, a batch size, a maximum iteration number and a node random discarding ratio; the deep learning neural network adopts four hidden layers; the activation function of each layer adopts a Selu function; a fourth hidden layer adopts a node random discarding mechanism;
the method for defining the loss function and the iterative optimization algorithm comprises the following steps: the loss function adopts mean square error MSE, the accuracy adopts mean error MAE, and the specific calculation formula is as follows:
wherein z is i Is an actual value, y i Is a predicted value.
Compared with the prior art, the invention has the advantages and beneficial effects that: the invention adopts the watershed socioeconomic classification grade data such as land utilization area, GDP, population and the like to predict the water quality in a watershed scale large range, overcomes the defects of small calculation range and low calculation speed caused by depending on numerical simulation in the traditional water quality prediction, simultaneously avoids the strict requirements of the traditional statistical method and the current common recurrent neural network on the continuity of input data for water quality prediction, can calculate the quantitative relation between the discrete socioeconomic classification grade data and the continuously changed water quality data more conveniently and improves the prediction practicability. In addition, the invention avoids the problem of social economic statistical data loss when adopting social economic data to predict the water quality, can predict only by adopting social economic classification grade data without accurate social economic data, has definite theoretical significance and simple and easy operation, and is easy to be applied in actual water quality management.
Drawings
FIG. 1 is a flow chart of the rapid watershed water quality prediction method of the invention;
FIG. 2 is a scatter plot of predicted values and measured values of CODMn;
FIG. 3 is a spatial distribution diagram of predicted values of CODMn at a certain day.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.
The technical solution of the present invention is further described below with reference to fig. 1 to 3 and the embodiment.
A water quality rapid prediction method based on continuous and graded mixed data is shown in figure 1 and comprises the following steps:
s1, determining a drainage basin range and internal partitions thereof, wherein the internal partitions comprise sub-drainage basin partitions and administrative partitions;
according to Chinese water resource partition of 'national comprehensive planning of water resources', a Zhujiang district is selected as an embodiment river basin. And (3) extracting the digital elevation data of the drainage basin of the embodiment and calculating the catchment area by adopting SRTM 90-meter digital elevation data of NASA to obtain 138 sub-drainage basins of the Zhujiang river film area. Extracting 75 geographical administrative regions in the Zhujiang district by adopting a national geographical information center (http:// www. Ngcc. Cn/ngcc /) national geographical region distribution diagram (SHP format); and drawing a grade city distribution map of the Zhujiang photo area.
S2, determining meteorological parameters of the drainage basin in a prediction period, wherein the meteorological parameters comprise rainfall, evaporation and air temperature data observed at meteorological point i at time t, and the data are respectively used
Representing; the meteorological parameters are then converted into sub-basin meteorological data. The method specifically comprises the following steps:
s21, extracting daily average rainfall, evaporation and air temperature data of 69 meteorological sites from 1 month 1 day in 2020 to 4 months 30 days in 2021 in the basin in the embodiment by adopting a Chinese ground climate data daily value data set (V3.0) provided by a Chinese meteorological data network;
s22, according to longitude and latitude coordinates of the meteorological sites, manufacturing a meteorological site space distribution map by adopting ARCGIS software, and importing rainfall, evaporation and air temperature data of each site and each time period into an attribute table of the site distribution map;
s23, taking the Zhujiang film area as a boundary, and performing spatial interpolation on the daily meteorological data of the station by adopting a Kriging interpolation method; and (4) making a batch processing program of a kriging interpolation method, generating raster data of rainfall, evaporation and air temperature of the Zhujiang piece area in each time period (day), and storing the raster data into a geographic space database. Wherein, rainfall, evaporation and air temperature data are 486 grid graphs respectively, one graph every day.
And S24, making a grid data processing program, taking the sub-watersheds as boundaries, extracting rainfall, evaporation and air temperature data of each sub-watershed of the Zhujiang film area in each time period (day), and storing the data into an EXCEL document. One EXCEL document per sub-basin, for a total of 138 EXCEL documents, each containing data for 3 variables (rainfall, evaporation, air temperature) for 486 periods (486 rows, 3 columns).
S3, determining the social and economic parameters of the sub-watersheds in the prediction time period, and generating a social and economic parameter set of the sub-watersheds; the socioeconomic parameters comprise land utilization area, frequent population, GDP per capita, population density, grain yield and municipal sewage treatment rate. The land utilization area comprises cultivated land, forest land, grassland, water areas, urban and residential land and the utilization area of unused land. The step S3 specifically includes the following steps:
s31, obtaining social and economic parameters of land utilization area, constant population, GDP, per-capita GDP, population density, grain yield and urban sewage treatment rate in a prediction period; more specifically: the above parameters can be obtained by: acquiring annual land utilization data of the whole watershed range in a prediction period by adopting satellite remote sensing data; acquiring the permanent population, GDP and grain yield of each year and each administrative district in a prediction period through regional statistical yearbook; and acquiring the per-year and per-administrative region GDP, population density and urban sewage treatment rate in the prediction time period through regional statistics yearbook.
S32, converting the social and economic parameters in the S31 into related parameters of the sub-watershed;
(1) Determining land utilization parameters of each sub-basin and each time interval, specifically: reclassifying the watershed land utilization data according to cultivated land, forest land, grassland, water areas, cities, residential land and unused land to generate grid data of the watershed land utilization. Then, the area of the six soil utilization types in each sub-basin is calculated according to the sub-basin partition, as shown in detail below.
Wherein j represents a sub-basin number; d denotes a land use type number including cultivated land (d = 1), woodland (d = 2), grassland (d = 3), water area (d = 4), city and residential land (d = 5), unused land (d = 6); t represents a time period (day), y represents a year; area is the area of a grid within the sub-basin j;
representing the number of grids belonging to the land of the d category in the sub-basin j in the y year;
And respectively representing the areas of the d types of land in the sub-basin j at the y-th year and the t-th time. Wherein time t is within the y year.
(2) Determining the standing population, GDP and grain yield of the drainage basin in a prediction period, specifically: and converting the social and economic data of the administrative region into relevant parameters of the sub-watershed by adopting an area weighted average method. Specifically, the following formula is given.
Wherein k represents the number of administrative districts in the drainage basin, and N is the total number of administrative districts in the drainage basin; j is the sub-basin number; e denotes the number of the socioeconomic parameter, e =1 denotes the standing population,e =2 for GDP and e =3 for grain yield;
the population, GDP and food yield of the everlasting population of the y year and the k administrative district;
And &>
Respectively the population of the permanent living, GDP and grain yield of the jth sub-basin at the y year and the t moment; wherein time t is within the y-th year.
The ratio of the area of the kth administrative district in the jth sub-basin to the total area of the kth administrative district is represented;
Indicating the area of the kth administrative district inside the jth sub-basin.
(3) Determining the per-capita GDP, population density and urban sewage treatment rate of the drainage basin in a prediction time period, specifically: the average population GDP, population density and the urban sewage treatment rate belong to proportional data, so the area weighted average method cannot be adopted for calculating parameters of the sub-basin. And calculating the per capita GDP, population density and urban sewage treatment rate of the sub-drainage basin by constructing a topological relation between the administrative region and the sub-drainage basin and by an arithmetic mean method. The details are as follows.
Wherein k represents the number of administrative districts in the drainage basin, and N is the total number of administrative districts in the drainage basin; j is the sub-basin number; a represents the number of the socioeconomic parameter,a =1 represents the per-capita GDP, a =2 represents the population density, and a =3 represents the municipal sewage treatment rate;
GDP for the average population in the y year and the k administrative district, population density and municipal sewage treatment rate.
And
the average GDP of people in the y year, the t moment and the j sub-basin, the population density and the urban sewage treatment rate are respectively; wherein time t is within the y year.
According to the above (steps S31 and S32), the steps S31 and S32 are more specifically: and extracting land utilization data of two years in the Zhujiang film areas 2020 and 2021 by adopting land utilization year data with the resolution of 1km provided by the Landsat 8 satellite, wherein 2 raster data graphs are obtained in total. A land utilization data processing program is compiled, and the sub-watershed of the Zhujiang river slide area 138 is taken as a range, and the area of each sub-watershed, cultivated land, forest land, grassland, water area, city and large-consumption and unused land is extracted; converting the annual land use data of the sub-watershed into day data, wherein the land area of each type of each day is equal to the corresponding land use data of the year of the day. Five types of land area data of the sub-watershed per day are stored into the EXCEL document. One EXCEL document per sub-basin, for a total of 138 EXCEL documents, each document containing data for 6 variables (six types of land area data) for 486 periods (486 rows, 6 columns).
And (4) collecting the permanent population, GDP, average population GDP, population density, grain yield and urban sewage treatment rate of 2020 years, 2021 years and 75 land-level municipal administrative regions in the Zhujiang film region by adopting a Chinese informed network yearbook database. And compiling a socioeconomic data processing program, and converting the socioeconomic data of 75 prefectures into the socioeconomic data of 138 sub-watersheds according to an area weighted average method and an arithmetic average method and the space topological relation between the prefecture administrative regions of the prefecture and the sub watersheds. And converting the annual socioeconomic data of the sub-watershed into daily data, wherein the daily socioeconomic data are equal to the corresponding data of the year of the day. The data is stored to the EXCEL document. One EXCEL document per sub-basin, 138 EXCEL documents, each containing 486 periods of 6 variable (6 socio-economic parameters) data (486 rows, 6 columns).
And S33, generating a social and economic parameter set of the sub-watershed.
S4, determining water quality parameters of the sub-watershed; the water quality parameters comprise concentration values of dissolved oxygen, COD (chemical oxygen demand) manganese, ammonia nitrogen, total phosphorus and total nitrogen of surface water; the method comprises the steps of adopting national water quality monitoring data provided by a national surface water quality automatic monitoring real-time data distribution system of a China environmental monitoring central office to extract daily average water quality data of 204 water quality sites from 1 month and 1 day in 2020 to 4 months and 30 days in 2021 in the Zhujiang district. The step S4 specifically includes the following steps:
s41, determining longitude and latitude coordinates of river channel water quality monitoring sites in the whole drainage basin (all sub-drainage basins), manufacturing a spatial distribution map of the water quality monitoring sites by adopting ARCGIS software, and then importing water quality index data of each river channel water quality site and each time period into an attribute table of the site distribution map.
S42, calculating water quality distribution in the river channels of the sub-watersheds;
because the surface water quality belongs to the parameters of the river water body, certain spatial continuity is provided in the river water body, but no spatial correlation on the river basin surface is provided, the water quality of the sub-river basin is calculated by adopting a river course line interpolation method instead of adopting the spatial interpolation method of the river basin surface. The method comprises the following specific steps:
(1) Collecting river network data (SHP data) of rivers above five levels in the whole river basin;
(2) And (3) with the river network of the river basin as a boundary, performing spatial interpolation on the water quality data of each time interval of the station by adopting an inverse distance weighting method to generate water quality grid data of each time interval along the river, and storing the water quality grid data into a geospatial database.
S43, taking the sub-watersheds as ranges, and calculating the average value of the river water quality in each sub-watersheds range to serve as the water quality parameter of the sub-watersheds. The calculation formula is as follows.
Wherein j represents the number of the sub-basin, t represents the time period, and s is the number of the grid of the river channel position in the sub-basin j; w is the number of the water quality parameter, w =1 represents dissolved oxygen, w =2 represents COD manganese, w =3 represents ammonia nitrogen, w =4 represents total phosphorus, w =5 represents total nitrogen;
and (3) representing water quality data of a grid s at the position of a river channel in the range of the sub-basin j.
Represents the total number of the grids in the river channel within the sub-basin j range>
And (3) water quality data of the sub-basin j at the time t is shown. />
One EXCEL document per sub-basin, for a total of 138 EXCEL documents, each document containing data of 5 variables (5 water quality indicators) for 486 periods (486 rows, 5 columns).
S5, carrying out grade classification on data in the social and economic parameter set of the sub-basin;
the data in the socio-economic parameter set of each sub-basin includes: cultivated land, woodland, grassland, water area, city and residential land, area of unused land, population of permanent dwellings, GDP and grain yield, GDP per capita, population density and urban sewage treatment rate, and the total is 12 variables. Setting a grade threshold value for each type of socio-economic variable, and calculating the grade value of each socio-economic variable as shown in the following formula.
Wherein j represents a sub-basin number, and t represents a time period; d represents a land use type number, d =1 represents cultivated land, d =2 represents forest land, d =3 represents grass land, d =4 represents water area, d =5 represents city and residential land, d =6 represents unused land; e represents the number of the socioeconomic parameter, e =1 represents the population of the living, e =2 represents GDP, e =3 represents the yield of the food; a represents the number of the socioeconomic parameter, a =1 represents the per-capita GDP, a =2 represents the population density, and a =3 represents the municipal wastewater treatment rate.
Showing the area of the d type land in the sub-basin j at the t moment;
A classification threshold representing a d category land area, N representing a number of classifications;
and (4) a grade value representing the land area of the class d in the sub-basin j at the t-th time.
A numerical value representing the e category social economic variable in the sub-basin j at the t moment;
A classification threshold representing a socioeconomic variable of the e category, L representing a number of classifications;
and (4) indicating the grade value of the e-type socioeconomic variable in the sub-basin j at the t-th time.
A numerical value representing a category a social economic variable in the sub-basin j at the t moment;
A classification threshold value representing a socioeconomic variable of the category a, and V represents a classification number;
And (4) indicating the grade value of the socioeconomic variable of the class a in the sub-basin j at the time t.
The classification is carried out by referring to the maximum and minimum values of socioeconomic variables of each sub-drainage basin and each category, the classification condition is that cultivated land is divided into 4 grades according to the area, forest land is divided into 5 grades, grassland is divided into 5 grades, water area is divided into 3 grades, city and residential land are divided into 3 grades, the area of unused land is divided into 3 grades, the standing population is divided into 4 grades, GDP and grain yield are divided into 4 grades, and the per capita GDP, population density and urban sewage treatment rate are divided into 3 grades.
S6, carrying out thermal coding on the social and economic parameters of the classification level to form a thermal coding matrix of the social and economic parameters;
s61, changing the socioeconomic grade variable of each sub-basin and each time interval
Converting into 0-1 row vector; in the row vector, the value of the nth column corresponding to the level value n of the social and economic level variable is 1; the values of the other elements in the row vector are 0; specifically, the formula is shown as follows:
ca 1，n ＝1 (12)
wherein j represents a sub-basin number, and t represents a time period; d represents a land use type number;
a grade value representing the d-type land area in the sub-basin j at the t moment is equal to n;
0-1 row vector, ca, corresponding to the level value of the land area of class d 1，n The value of the nth column in the row vector is 1.N represents the total number of levels of the d categories of land areas. Using the method described above, other socio-economic class variables are calculated>
And &>
0-1 row vectors of (a).
S62, constructing a thermal coding 0-1 matrix of the socioeconomic grade variable.
0-1 row vector corresponding to grade values of population, GDP and grain yield of standing population
0-1 line vector corresponding to the grade values of human average GDP, population density and municipal sewage treatment rate>
Combined together to form a matrix>
Specifically, the following formula is shown.
S7, for meteorological parameters
And a water quality parameter->
Carrying out standardization treatment; specifically, the formula is shown as follows:
wherein,
weather and water quality parameters representing the sub-basin j at the t-th moment, including->
And &>
And &>
At all times t and t for each index respectivelyMaximum and minimum values in all sub-basin j;
Is normalized data.
s8, dividing a training set and a testing set in a thermal coding matrix of social and economic parameters, standardized sub-basin meteorological data and water quality parameters;
combining the heat coding 0-1 matrixes of the continuous sub-watershed meteorological data and the social economic data into a data matrix of t x j rows and 3+ N + L + V columns as an input digital matrix for deep learning, and representing the data matrix by TableX; converting the continuous water quality data into a data matrix of t multiplied by j rows to serve as an output label matrix of deep learning, and expressing the output label matrix by TableY; then, 70% of rows in the input digital matrix and the output label matrix are used as a training set, and 30% of rows are used as a test set, and the following specific formula is shown as follows:
more specifically, the meteorological parameters of the sub-watersheds and the socioeconomic 0-1 matrix parameters are integrated into one data set. The data is stored in EXCEL tables, one EXCEL document per sub-domain, for a total of 138 EXCEL documents. Each document contains 44 columns of data for 486 periods, with the variables including: rainfall, evaporation, air temperature continuity variables, and 0-1 row vector of grade data of cultivated land, forest land, grassland, water areas, cities and residents, area of unused land, population of standing living, GDP, population average GDP, population density, grain yield, and municipal sewage treatment rate. The 138 EXCEL document data are merged into 1 data matrix. This matrix is designated as TableX and used as an input data matrix for deep learning.
And integrating the water quality parameters of the sub-watersheds into a data set. And combining the water quality data of the 138 sub-watersheds into 1 data matrix. The matrix column is 5 water quality variables. The rows of the matrix are 138 sub-watersheds and 486 days, and the total number is 67068 rows. This matrix is denoted as TableY and is used as the output label matrix for deep learning.
TableX and TableY are normalized and processed with the largest and smallest normalization. Each data in the matrix is subtracted by the minimum value of the column in which the data is located, divided by the difference between the maximum and minimum values of the column. Dividing the matrixes TableX and TableY into a training set and a testing set, and taking 70% of rows as the training set to form TrainX and TrainY;30% of the rows are taken as a test set, resulting in TestX and TestY.
S9, constructing a fully connected deep learning neural network (DNN), and defining a loss function and an iterative optimization algorithm;
the deep learning neural network parameters comprise the number k of network hidden layers, the number n of nodes of each hidden layer, an activation function, a loss function (Ioss), an accuracy calculation function (metrics), a learning rate (learning rate), a batch size (batch size), a maximum iteration number (epochs) and a node random discard rate (dropout). The network adopts four hidden layers, namely 128 nodes in the first layer, 64 nodes in the second layer, 32 nodes in the third layer and 16 nodes in the fourth layer; in the embodiment, the discard rate is 20%, the learning rate (learning rate) is 0.001, the volume size (batch size) is 64, and the maximum iteration number (epochs) is 1000.
The defining loss function and the iterative optimization algorithm are specifically as follows: the loss function uses Mean Squared Error (MSE) and the accuracy uses Mean Absolute Error (MAE). The calculation formula is shown in the following formula. The training process is optimized by adopting an Adam algorithm.
Wherein z is i Is the actual value, y i Is a predicted value. The smaller the MSE and MAE values are, the closer the predicted value and the true value of the model are, the higher the precision of the model is, and the better the performance is.
S10, inputting the training set into a deep learning neural network to obtain a trained deep learning neural network, and then inputting the test set into the trained deep learning neural network to obtain a predicted water quality parameter; comparing the predicted water quality parameters with the actually measured water quality parameters, adjusting model parameters, and finally storing the deep learning neural network and parameter data thereof which meet the precision requirement;
more specifically: the DNN network is trained using TrainX as an input of the DNN network and TrainY as an output label. The weights of the DNN network are iteratively updated based on the training data until the model converges. And adopting the trained DNN network to predict the water quality.
And inputting the TestX into the DNN network, and predicting to obtain the corresponding water quality parameter. And performing denormalization on the predicted value to obtain an actual value, and denormalizing the TestY to obtain a water quality measured value. Then, a scatter diagram of the predicted actual value and the actually measured water quality value is drawn, and the square of the correlation coefficient r (r) is calculated 2 )，r 2 Larger indicates more accurate prediction results. Continuously comparing the predicted value with the measured value, and calculating corresponding MSE and r 2 According to MSE and r 2 And adjusting the model parameters, and re-training to ensure the accuracy. The DNN network meeting the precision requirement and the parameter data such as the weight of the DNN network are stored, and decision support is provided for management personnel. The denormalization formula is as follows:
y′ i ＝y min +y i ·(y max -y min ) (20)
wherein, y' i The predicted value after reduction; y is i Is a predicted value output by the DNN network.
S11, selecting meteorological and social economic data of a certain time period of the drainage basin, inputting the meteorological and social economic data into the deep learning neural network stored in S10, and predicting to obtain water quality parameter values of all sub drainage basins in the time period, wherein the water quality parameter values comprise concentration values of dissolved oxygen, COD (chemical oxygen demand) manganese, ammonia nitrogen, total phosphorus and total nitrogen. And drawing a spatial distribution diagram of the water quality of the sub-basin for each water quality index, and converting the distribution diagram into a grid data format. And taking the river network of the river basin as a boundary, and extracting water quality index data of the grid at the position of the river channel to obtain a water quality distribution map of the river network. In the present embodiment, the spatial distribution of the predicted value and the actual measured value of CODMn and the predicted value of CODMn on a certain day are shown in fig. 2 and fig. 3, respectively.
Compared with the prior art, the beneficial effects of this embodiment are: in the embodiment, the Zhujiang river basin is taken as an example, the socioeconomic classification grade data of land utilization area, population density, grain yield and urban sewage treatment rate are adopted to predict the water quality in the large scale of the basin, the defects of small calculation range and low calculation speed caused by numerical simulation in the traditional water quality prediction are overcome, the strict requirements of the traditional statistical method and the current common recurrent neural network on the continuity of input data in the water quality prediction are avoided, the quantitative relation between the discrete socioeconomic classification grade data and the continuously changed water quality data can be calculated more conveniently, and the prediction practicability is improved. In addition, the invention avoids the problem of social and economic statistical data loss when the water quality is predicted by adopting the social and economic data, can predict by only adopting the social and economic classification grade data without accurate social and economic data, has definite theoretical significance and simple and easy operation, and is easy to apply to actual water quality management.
It should be noted that the above-mentioned embodiments are only used for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present invention can be modified or substituted with equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (10)
1. A water quality rapid prediction method based on continuous and graded mixed data is characterized by comprising the following steps:
s1, determining a drainage basin range and an internal partition thereof, wherein the internal partition comprises a sub-drainage basin partition and an administrative partition;
s2, determining meteorological parameters of the drainage basin in a prediction time period, and converting the meteorological parameters into sub-drainage basin meteorological data; the meteorological parameters comprise rainfall, evaporation and air temperature data observed at meteorological points at time i;
s3, determining the social and economic parameters of the sub-watersheds in the prediction time period, and generating a social and economic parameter set of the sub-watersheds;
s4, determining water quality parameters of the sub-basin;
s5, carrying out grade classification on data in the social and economic parameter set of the sub-basin; calculating the grades of various socioeconomic variables according to the grade threshold of the socioeconomic parameters to obtain the grade classification of high, medium and low socioeconomic development degree of the sub-basin;
s6, carrying out thermal coding on the socioeconomic parameters of the grade classification to form a thermal coding matrix of the socioeconomic parameters;
coding the grade of the social and economic parameters in a form of 0-1 row vector, wherein the length of the row vector is the total number N of grades of the social and economic variables, namely the row vector is 1 row and N columns; in the row vector, the value of the nth column corresponding to the social and economic grade variable grade value n is 1; the values of the other elements in the row vector are 0; splicing the level hot coded row vectors of different socioeconomic variables of a certain sub-basin left and right according to columns to obtain the level hot coded row vectors of all socioeconomic variables of a certain sub-basin;
stacking the level thermal coding row vectors of the social and economic variables of all the sub-watersheds up and down according to rows to obtain a thermal coding matrix of the social and economic variable levels of all the watersheds;
s7, carrying out standardization treatment on the meteorological parameters and the water quality parameters;
s8, dividing a training set and a testing set in a thermal coding matrix of social and economic parameters, standardized sub-basin meteorological data and water quality parameters; splicing the continuous sub-watershed meteorological data and the discrete socioeconomic grade data of the 0-1 matrix according to columns to obtain a unified digital matrix which is used as an input digital matrix of a deep learning neural network to realize the unified processing of the continuous data and the discrete data;
s9, constructing a full-connection deep learning neural network, and defining a loss function and an iterative optimization algorithm;
s10, inputting the training set into a deep learning neural network to obtain a trained deep learning neural network, and then inputting the test set into the trained deep learning neural network to obtain a predicted water quality parameter; comparing the predicted water quality parameters with the actually measured water quality parameters, adjusting model parameters, and finally storing the deep learning neural network and parameter data thereof which meet the precision requirement;
s11, selecting meteorological and social economic data of a certain time period of the drainage basin, inputting the meteorological and social economic data into the deep learning neural network stored in the step S10, and predicting to obtain water quality parameters of all the sub-drainage basins in the time period.
2. The method of claim 1, wherein: step S2 is more specifically:
s21, acquiring rainfall, evaporation and air temperature data in a prediction time period by a meteorological site;
s22, manufacturing a meteorological site space distribution diagram according to longitude and latitude coordinates of meteorological sites, and importing rainfall, evaporation and air temperature data of each site and each time period into an attribute table of the site distribution diagram;
s23, with the whole drainage basin as a boundary, performing spatial interpolation on meteorological data of each station by adopting a Krigin interpolation method to generate grid data of rainfall, evaporation and air temperature of the drainage basin in each period, and storing the grid data into a geospatial database;
and S24, calculating the average value of rainfall, evaporation and air temperature data of all grids in each sub-basin range by taking the sub-basins as ranges to obtain meteorological parameters of the sub-basins.
3. The method of claim 1, wherein: the socioeconomic parameters comprise land utilization area, frequent population, GDP per capita, population density, grain yield and urban sewage treatment rate.
4. The method of claim 3, wherein: step S3 is more specifically:
s31, obtaining social and economic parameters of land utilization area, constant population, GDP, per-capita GDP, population density, grain yield and urban sewage treatment rate in a prediction period;
s32, converting the social and economic parameters in the step S31 into related parameters of the sub-watershed;
and S33, generating a socioeconomic parameter set of the sub-watershed.
5. The method of claim 1, wherein: the water quality parameters comprise concentration values of dissolved oxygen, COD (chemical oxygen demand) manganese, ammonia nitrogen, total phosphorus and total nitrogen of surface water; step S4 is more specifically:
s41, determining longitude and latitude coordinates of the riverway water quality monitoring stations in all the sub-flow domains to manufacture a space distribution diagram of the water quality monitoring stations, and then importing the water quality index data of each riverway water quality monitoring station and each time period into an attribute table of the station distribution diagram;
s42, calculating water quality distribution in the river channel of the sub-basin;
s43, taking the sub-watersheds as ranges, and calculating the average value of the river water quality in each sub-watersheds range to serve as the water quality parameter of the sub-watersheds.
6. The method of claim 5, wherein: step S42 is more specifically:
s421, collecting river network data of rivers with more than five levels in the whole river basin;
and S422, with the river network of the river basin as a boundary, performing spatial interpolation on the water quality data of each time interval of the station by adopting an inverse distance weighting method to generate water quality grid data of each time interval along the river, and storing the water quality grid data into a geospatial database.
7. The method of claim 4, wherein: the land utilization area comprises cultivated land, forest land, grassland, water areas, urban and residential land and the utilization area of unused land; step S5 is more specifically: calculating grade values of the socioeconomic variables to form socioeconomic grade variables, wherein the grade values are the areas of cultivated land, forest land, grassland, water area, city and residential land, unused land, and the grade thresholds of socioeconomic parameters of the permanent population, GDP, grain yield, per capita GDP, population density and urban sewage treatment rate.
8. The method of claim 7, wherein: step S6 is more specifically:
s61, converting the socioeconomic grade variable of each sub-basin and each time interval into 0-1 row vectors; the length of the row vector is the total number N of the socioeconomic variables, namely the row vector is 1 row and N columns; in the row vector, the value of the nth column corresponding to the level value n of the social and economic level variable is 1; the values of the other elements in the row vector are 0;
s62, constructing a thermal coding 0-1 matrix of the socioeconomic grade variable.
9. The method of claim 8, wherein: step S8 is more specifically: combining the thermal coding 0-1 matrixes of the sub-basin meteorological data and the social and economic data into a digital matrix which is used as an input digital matrix of the deep learning neural network; converting the sub-basin water quality parameters into a digital matrix as an output label matrix of the deep learning neural network; then 70% of the rows in the input digit matrix and the output label matrix are used as a training set, and 30% of the rows are used as a testing set.
10. The method of claim 1, wherein: the parameters of the deep learning neural network comprise the number k of hidden layers of the network, the number n of nodes of each hidden layer, an activation function, a loss function, an accuracy rate calculation function, a learning rate, a batch size, the maximum iteration number and a node random discarding rate; the deep learning neural network adopts four hidden layers; the activation function of each layer adopts a Selu function; a node random discarding mechanism is adopted in the fourth hidden layer;
the method for defining the loss function and the iterative optimization algorithm comprises the following steps: the loss function adopts mean square error MSE, the accuracy adopts mean error MAE, and the specific calculation formula is as follows:
wherein z is i Is an actual value, y i Is a predicted value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111044820.7A CN113723704B (en) | 2021-09-07 | 2021-09-07 | Water quality rapid prediction method based on continuous and graded mixed data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111044820.7A CN113723704B (en) | 2021-09-07 | 2021-09-07 | Water quality rapid prediction method based on continuous and graded mixed data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113723704A CN113723704A (en) | 2021-11-30 |
| CN113723704B true CN113723704B (en) | 2023-04-18 |
Family
ID=78682340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111044820.7A Active CN113723704B (en) | 2021-09-07 | 2021-09-07 | Water quality rapid prediction method based on continuous and graded mixed data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113723704B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109740680A (en) * | 2019-01-07 | 2019-05-10 | 深圳中创华安科技有限公司 | A kind of classification method and system of mixing value attribute examination & approval data |
| CN111387938A (en) * | 2020-02-04 | 2020-07-10 | 华东理工大学 | Patient heart failure death risk prediction system based on feature rearrangement one-dimensional convolutional neural network |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2002337963A1 (en) * | 2001-10-22 | 2003-05-06 | Emery, Coppola, J. , Jr. | Neural network based predication and optimization for groundwater / surface water system |
| CN109242203A (en) * | 2018-09-30 | 2019-01-18 | 中冶华天南京工程技术有限公司 | A kind of water quality prediction of river and water quality impact factors assessment method |
| US11544530B2 (en) * | 2018-10-29 | 2023-01-03 | Nec Corporation | Self-attentive attributed network embedding |
| CN112529234A (en) * | 2019-09-18 | 2021-03-19 | 上海交通大学 | Surface water quality prediction method based on deep learning |
| CN112633645B (en) * | 2020-12-10 | 2023-05-02 | 长江水利委员会长江科学院 | Social and economic benefit accounting method for water resource mining and efficient utilization effects of river source arid region |
| CN113011992B (en) * | 2021-03-19 | 2024-03-15 | 中国水利水电科学研究院 | River basin agricultural non-point source pollution river entering coefficient measuring and calculating method based on standard data |
-
2021
- 2021-09-07 CN CN202111044820.7A patent/CN113723704B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109740680A (en) * | 2019-01-07 | 2019-05-10 | 深圳中创华安科技有限公司 | A kind of classification method and system of mixing value attribute examination & approval data |
| CN111387938A (en) * | 2020-02-04 | 2020-07-10 | 华东理工大学 | Patient heart failure death risk prediction system based on feature rearrangement one-dimensional convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113723704A (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Setegn et al. | Spatial delineation of soil erosion vulnerability in the Lake Tana Basin, Ethiopia | |
| CN113723703A (en) | Water quality prediction method and system based on multi-source data fusion and deep learning | |
| CN107292098A (en) | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology | |
| Liu et al. | Optimal allocation of water quantity and waste load in the Northwest Pearl River Delta, China | |
| KR20170005553A (en) | Floods, drought assessment and forecasting techniques development for intelligent service | |
| Kisi et al. | Estimation of long-term monthly temperatures by three different adaptive neuro-fuzzy approaches using geographical inputs | |
| Das et al. | FORWARD: a model for forecasting reservoir water dynamics using spatial Bayesian network (SpaBN) | |
| Kim et al. | Bias correction of daily precipitation over South Korea from the long-term reanalysis using a composite Gamma-Pareto distribution approach | |
| CN111428936B (en) | River basin rainfall flood availability index measuring and calculating method based on distributed water nodes | |
| Li et al. | A method of rainfall runoff forecasting based on deep convolution neural networks | |
| CN116205136A (en) | Large-scale river basin deep learning flood forecasting method based on runoff lag information | |
| Tursunova et al. | Water resources of Kazakhstan in conditions of uncertainty | |
| US20240256746A1 (en) | Scheduling method and system for operation of reservoirs to recharge freshwater for repelling saltwater intrusion under changing conditions | |
| Leela Krishna et al. | Optimal multipurpose reservoir operation planning using Genetic Algorithm and Non Linear Programming (GA-NLP) hybrid approach | |
| CN115496148A (en) | Method and device for evaluating large-scale hydrological situation change of river | |
| CN110110391A (en) | A kind of flood estimation method in the scarce measuring runoff data basin based on region division | |
| Herath et al. | Evaluation of HEC-HMS model for water resources management in Maha Oya Basin in Sri Lanka | |
| CN118095034B (en) | Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity | |
| Patel et al. | Rainfall Prediction using Machine Learning Techniques for Sabarmati River Basin, Gujarat, India. | |
| CN117172037B (en) | Distributed hydrologic forecasting method, device, computer equipment and medium | |
| Xia et al. | Urban water system theory and its model development and application | |
| CN113762615A (en) | Flood prediction method and device, computer equipment and storage medium | |
| Aminjavaheri et al. | Determining the robust optimal set of BMPs for urban runoff management in data-poor catchments | |
| CN113723704B (en) | Water quality rapid prediction method based on continuous and graded mixed data | |
| Diodato et al. | Case study for investigating groundwater and the future of mountain spring discharges in Southern Italy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |