CN118095034A - Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity - Google Patents
Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity Download PDFInfo
- Publication number
- CN118095034A CN118095034A CN202311518460.9A CN202311518460A CN118095034A CN 118095034 A CN118095034 A CN 118095034A CN 202311518460 A CN202311518460 A CN 202311518460A CN 118095034 A CN118095034 A CN 118095034A
- Authority
- CN
- China
- Prior art keywords
- water
- model
- taking
- hydrologic
- lstm
- 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.)
- Granted
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 436
- 238000010168 coupling process Methods 0.000 title claims abstract description 84
- 230000008878 coupling Effects 0.000 title claims abstract description 82
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 82
- 230000000694 effects Effects 0.000 title claims abstract description 55
- 230000009977 dual effect Effects 0.000 title claims abstract description 27
- 238000013277 forecasting method Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 59
- 238000004088 simulation Methods 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000005096 rolling process Methods 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims description 33
- 239000002689 soil Substances 0.000 claims description 20
- 238000004519 manufacturing process Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 11
- 241000282414 Homo sapiens Species 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011161 development Methods 0.000 claims description 7
- 230000018109 developmental process Effects 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 230000002262 irrigation Effects 0.000 claims description 5
- 238000003973 irrigation Methods 0.000 claims description 5
- 239000002352 surface water Substances 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 2
- 239000008239 natural water Substances 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000003673 groundwater Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- 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
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Water Supply & Treatment (AREA)
- Mathematical Physics (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a hydrologic forecasting method and a hydrologic forecasting system based on hydrologic model and LSTM dual coupling under the influence of water taking activities, which can realize accurate and reliable simulation of a social-natural hydrologic cycle coupling process and accurately forecast water taking quantity and hydrologic variables. The method comprises the following steps: step 1, constructing a distributed hydrological model according to the range of a forecast river basin; step 2, the water users are in one-to-one correspondence with the grid cells and the sub-basins in the hydrological model; step 3, performing space-time interpolation on the water usage influence factors, and performing space interpolation on the stream domain attribute data to obtain all water usage influence factor data corresponding to each grid; step 4, taking the mean value of the influence factors as the input of the LSTM, taking the water consumption of the water user as a target value, and training the LSTM; and 5, constructing a coupling forecasting model by coupling the LSTM model and the distributed hydrologic model, and carrying out rolling simulation forecasting on the water consumption and hydrologic variables.
Description
Technical Field
The invention belongs to the technical field of hydrologic simulation and forecasting, and particularly relates to a hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under the influence of water taking activities.
Background
The watershed hydrologic process is influenced by both meteorological driving processes such as precipitation and radiation and artificial water taking processes such as water taking, water supplying, water using, water consuming and draining. Under the influence of the current strong human activities, the influence of the social water taking on the hydrologic process of the river basin is larger and larger, and the water resource flux caused by the artificial interference of a plurality of river basins in the north of China even exceeds the actual measurement flux of the natural process. In the social water taking process, various elements such as total water amount, process, space-time distribution and the like of different water sources and water taking main bodies have great uncertainty, and the drainage basin water resource system is deeply influenced in the aspects of circulating paths and circulating characteristics. The study of the hydrologic process of the watershed under the influence of the water taking activity of the society still faces a plurality of key scientific problems to be broken through and solved urgently.
Currently, there are two main methods for hydrologic simulation considering human water intake activities: (1) The statistical value input, namely, taking the total water amount input based on regional year statistics, matching the space-time scale of the water taking amount with the model calculation scale through uniform interpolation in time and space in the model, and taking the space-time scale of the water taking amount as the model input; (2) And estimating the population of the area and the total production value of the area by a quota method, counting the water consumption of the area based on the water consumption of the population unit and the total production value of the area, estimating the water consumption of the area under investigation by the quota method, and then inputting the water consumption as a hydrological model to construct a hydrological process simulation model under the action of water.
However, in recent years, as the sustainable development of water resources and the awareness of water ecological protection of society are gradually enhanced, the formulation of a social water consumption scheme is also influenced by the state of natural water resources in order to ensure the ecological flow of the downstream. Therefore, the total water consumption of the area is estimated only based on the socioeconomic conditions, the space-time difference of the water consumption and the state of natural water resources are ignored, and the requirement of the current hydrologic process simulation on the water consumption activity data of human beings cannot be met.
In most water consumption prediction researches, only a prediction model is utilized, the prediction model is built according to the internal law of water consumption historical data to predict, the adopted water consumption data are mostly in annual scale, the water consumption data cannot be combined with the actual influence factors of water consumption, objective law is not met, and meanwhile, the predicted space-time scale cannot meet application requirements. In the water using process, the water resource of the area is affected, and the water demand of the society cannot be satisfied in the area with lack of the water resource. The water resources are unevenly distributed, the water resources are excessively developed due to social development, and the climate change is caused, so that the available water quantity in some areas cannot meet the social water demand. The change of hydrologic climate can cause the change of short-term water resource quantity, and the social economic factor has a great influence on long-term water consumption.
Disclosure of Invention
The invention aims to solve the problems, and aims to provide a hydrologic forecasting method and a hydrologic forecasting system under the influence of water taking activities based on double coupling of a hydrologic model and LSTM, which fully consider the influence of the water taking activities on a natural hydrologic process and the influence of regional water resource quantity on actual water taking quantity, realize accurate and reliable simulation of a social-natural water circulation coupling process and accurately forecast water taking quantity and hydrologic variables.
In order to achieve the above object, the present invention adopts the following scheme:
< method >
The invention provides a hydrologic forecasting method based on hydrologic model and LSTM dual coupling under the influence of water taking activities, which comprises the following steps:
Step 1, constructing a distributed hydrological model according to the range of a forecast river basin, dividing the river basin into grid units, and dividing sub-river basins based on a digital water system; then, parameter calibration is carried out on the distributed hydrologic model according to the historical data;
step 2, according to the geographical positions of the water taking users, the water taking users are in one-to-one correspondence with grid cells and sub-watershed in the hydrologic model;
Step3, performing space-time interpolation on the water usage influence factors, and performing space interpolation on the stream domain attribute data to obtain all water usage influence factor data corresponding to each grid;
Step 4, after normalization processing is carried out on all the water taking influence factor data and the daily water taking quantity, selecting an influence factor mean value in the influence area range of each water taking user as an input of an LSTM model, taking the water taking quantity of the water taking user as a target value, and training the LSTM model;
step 5, coupling the LSTM model and the distributed hydrologic model to construct a coupling forecast model:
Based on the water taking influence factors, predicting the water taking amount through a trained LSTM model; then, inputting the predicted water consumption into a distributed hydrological model to predict the water resource quantity; the water resource quantity obtained by the prediction of the distributed hydrological model becomes a water taking influence factor of the LSTM model water taking quantity prediction of the next time step (the water taking influence factor data of the next time step is updated by the prediction of the distributed hydrological model); the rolling prediction is used for forming the mutual coupling feedback of the water taking quantity and the water resource, constructing a coupling prediction model under the influence of the water taking activity of human beings, and performing simulation prediction on the water taking quantity and the hydrologic variable by the prediction coupling model.
Preferably, in the method for forecasting hydrologic under the influence of water taking activity based on dual coupling of a hydrologic model and an LSTM, in step 5, in the operation process of the coupled forecasting model, firstly, based on water taking quantity at time t 0 and meteorological driving data of a current time step, the method simulates hydrologic conditions in a river basin, including Q and SM:
Qt＝φ(B,Mt,WWt)
SMt＝g(B,Mt,SMt-1,WWt)
wherein: phi and g respectively represent simulation processes of the model on runoff of a flow field and water content of soil under the driving of meteorological data, underlying surface data and water data;
Then, at the next time step, based on the already trained LSTM water taking prediction network f (X), the hydrologic condition H t simulated by the previous day distributed hydrologic model, and the meteorological factor M t+1, the socioeconomic factor E t+1 of the current time step, the water taking WW at time t+1 is predicted:
WWt+1＝f(B,Et,Mt,SMt,Qt)
Therefore, under the drive of meteorological data, underlying surface data and socioeconomic data, the coupling forecasting model carries out rolling simulation calculation on the water resource condition and the water taking quantity in the flow area, and the forecasting of the water taking quantity under the constraint of the water resource and the simulation of the runoff of the flow area under the influence of the water taking are realized.
In the method for forecasting the hydrologic under the influence of the water taking activity based on the dual coupling of the hydrologic model and the LSTM, in the step 5, the influence factor data of the day of the forecast day and the daily water taking quantity of the three days before the forecast day are input into the trained LSTM model after normalization processing, the daily water taking quantity of the forecast day is forecast, and the result of the forecast water taking quantity is output to the distributed hydrologic model;
Taking the influence of water taking activity into consideration in the production and collection process of the distributed hydrologic model, dividing a water taking mode into surface water taking and underground water taking, dividing the water taking mode into agricultural water and non-agricultural water, and identifying the water taking process predicted by the LSTM model;
The input variables of the LSTM model comprise meteorological factors, drainage basin attribute factors, socioeconomic factors and water resource quantity factors, the meteorological factors, the drainage basin attribute factors and the socioeconomic factors are used as external inputs, drainage basin outlet runoffs Q and soil water contents SM simulated by the distributed hydrologic model are used as water resource quantity factor inputs and are jointly input into the LSTM model, and the water taking quantity WW t+1 of a user taken in the next time step is predicted based on a trained water taking prediction network.
Preferably, in the method for forecasting hydrologic under the influence of water taking activity based on dual coupling of a hydrologic model and an LSTM, the water taking influence factor includes: meteorological factors, socioeconomic factors, basin attribute factors, water history factors, and water resource quantity factors are taken.
Preferably, the method for forecasting hydrologic under the influence of taking water activities based on dual coupling of a hydrologic model and LSTM provided by the invention has weather factors of daily scale weather data, and comprises the following steps: rainfall, wind speed, temperature, relative humidity, and sunlight duration;
socioeconomic factors are socioeconomic developments, including: population density, domestic production total GDP, night light index, irrigation area;
the drainage basin attribute factor is drainage basin underlying data, comprising: geographic elevation DEM, land utilization type;
the taking water history factor is taking water user basic information, and comprises the following steps: taking a water taking source, a water type and a daily water taking amount monitoring value of a water user;
Water resource quantity factors, including runoff and soil moisture.
< System >
Furthermore, the invention also provides a hydrological forecasting system based on the hydrological model and LSTM dual coupling, which automatically realizes the method, and is characterized by comprising the following components:
A hydrological model construction unit which constructs a distributed hydrological model according to the range of the forecast basin, divides the basin into grid cells, and divides the sub-basin based on the digital water system; then, parameter calibration is carried out on the distributed hydrologic model according to the historical data;
The water taking user corresponding part is used for carrying out one-to-one correspondence on each water taking user and grid cells and sub-watershed in the hydrologic model according to the geographic position of the water taking user;
A water usage influence factor obtaining part for performing space-time interpolation on the water usage influence factors and performing spatial interpolation on the stream domain attribute data to obtain all water usage influence factor data corresponding to each grid;
The LSTM model training part is used for carrying out normalization processing on all the water taking influence factor data and the daily water taking quantity, selecting an influence factor mean value in the influence area range of each water taking user as the input of the LSTM model, taking the water taking quantity of the water taking user as a target value, and training the LSTM model;
A coupling model construction part for constructing a coupling forecast model by coupling the LSTM model and the distributed hydrologic model: based on the water taking influence factors, predicting the water taking amount through a trained LSTM model; then, inputting the predicted water consumption into a distributed hydrological model to predict the water resource quantity; the water resource quantity predicted by the distributed hydrologic model becomes a water taking influence factor of the LSTM model water taking quantity prediction of the next time step; the rolling prediction is used for forming the mutual coupling feedback of the water taking quantity and the water resource, constructing a coupling prediction model under the influence of the water taking activity of human beings, and performing simulation prediction on the water taking quantity and the hydrologic variable by the prediction coupling model;
The control part is communicated with the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part, the coupling model building part and the water taking influence factor based on the water taking influence factor, and controls the operation of the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part and the coupling model building part.
Preferably, the hydrologic forecasting system based on the hydrologic model and LSTM dual coupling under the influence of water taking activities provided by the invention can further comprise: and the input display part is in communication connection with the control part, allows a user to input an operation instruction, and displays input, output and intermediate processing data of the corresponding part in a text, form, graph, static or dynamic model mode according to the operation instruction.
Preferably, in the hydrologic forecasting system under the influence of the water taking activity based on the dual coupling of the hydrologic model and the LSTM, the coupling forecasting model simulates hydrologic conditions in a river basin based on the water taking amount at the time t 0 and meteorological driving data of the current time step in the operation process, and the hydrologic conditions comprise Q and SM:
Qt＝φ(B,Mt,WWt)
SMt＝g(B,Mt,SMt-1,WWt)
wherein: phi and g respectively represent simulation processes of the model on runoff of a flow field and water content of soil under the driving of meteorological data, underlying surface data and water data;
Then, at the next time step, based on the already trained LSTM water taking prediction network f (X), the hydrologic condition H t simulated by the previous day distributed hydrologic model, and the meteorological factor M t+1, the socioeconomic factor E t+1 of the current time step, the water taking WW at time t+1 is predicted:
WWt+1＝f(B,Et,Mt,SMt,Qt)
Therefore, under the drive of meteorological data, underlying surface data and socioeconomic data, the coupling forecasting model carries out rolling simulation calculation on the water resource condition and the water taking quantity in the flow area, and the forecasting of the water taking quantity under the constraint of the water resource and the simulation of the runoff of the flow area under the influence of the water taking are realized.
Preferably, in the hydrologic forecasting system under the influence of the water taking activity based on the dual coupling of the hydrologic model and the LSTM, in the coupled forecasting model, the influence factor data of the day of the forecast day and the daily water taking quantity of the three days before the forecast day are input into the trained LSTM model after normalization processing, the daily water taking quantity of the forecast day is forecast, and the result of the forecast water taking quantity is output to the distributed hydrologic model;
Taking the influence of water taking activity into consideration in the production and collection process of the distributed hydrologic model, dividing a water taking mode into surface water taking and underground water taking, dividing the water taking mode into agricultural water and non-agricultural water, and identifying the water taking process predicted by the LSTM model;
The input variables of the LSTM model comprise meteorological factors, drainage basin attribute factors, socioeconomic factors and water resource quantity factors, the meteorological factors, the drainage basin attribute factors and the socioeconomic factors are used as external inputs, drainage basin outlet runoffs Q and soil water contents SM simulated by the distributed hydrologic model are used as water resource quantity factor inputs and are jointly input into the LSTM model, and the water taking quantity WW t+1 of a user taken in the next time step is predicted based on a trained water taking prediction network.
Preferably, in the hydrologic forecasting system under the influence of the taking water activity based on the dual coupling of the hydrologic model and the LSTM, the taking water influence factor includes: meteorological factors, socioeconomic factors, basin attribute factors, water history factors and water resource quantity factors are taken;
The meteorological factors are day scale meteorological data comprising: rainfall, wind speed, temperature, relative humidity, and sunlight duration;
socioeconomic factors are socioeconomic developments, including: population density, domestic production total GDP, night light index, irrigation area;
the drainage basin attribute factor is drainage basin underlying data, comprising: geographic elevation DEM, land utilization type;
the taking water history factor is taking water user basic information, and comprises the following steps: taking a water taking source, a water type and a daily water taking amount monitoring value of a water user;
Water resource quantity factors, including runoff and soil moisture.
Effects and effects of the invention
According to the hydrologic forecasting method and system under the influence of the water taking activities based on the double coupling of the hydrologic model and the LSTM, according to the geographical positions of the water taking users, grid units in the water taking users and the hydrologic model are corresponding to the sub-watershed, then all water taking influence factor data corresponding to each grid are obtained, then the influence factor mean value in the influence area range of each water taking user is selected to serve as the input of the LSTM, the water taking amount of the water taking users is taken as a target value, and the LSTM model is trained; then, based on the taken water influence factors, predicting the taken water quantity through a trained LSTM model, inputting the predicted taken water quantity into a distributed hydrological model, predicting the water resource quantity by considering the taken water activity, updating the taken water influence factors of the next time step by using the water resource quantity predicted by the distributed hydrological model, and inputting the updated taken water influence factors into the LSTM model for taking water quantity prediction; the water consumption and runoff are predicted in a rolling way, and the mutual coupling feedback of the water consumption and the water resource is formed, so that the water consumption activity can be fully considered, the accurate and reliable simulation of the social-natural water circulation coupling is realized, and the simulation prediction result is more in line with the actual situation.
The invention can simulate and forecast the water resource condition and the water consumption in the river basin in future climate situations or in the areas with water shortage/water consumption data, so that the result of the hydrologic forecast in the areas with water shortage/water consumption data is more accurate. The invention not only can provide the forecast of the water taking in the future climate mode, but also can provide the hydrologic forecast result influenced by the water taking activity, can better provide support for the establishment of water resource management and scheduling schemes, and provides scientific tools for sustainable development of water resources, water ecological protection and reasonable guarantee of water demand.
Drawings
FIG. 1 is a flow chart of a method for hydrologic forecasting under the influence of water taking activities based on dual coupling of a hydrologic model and an LSTM according to an embodiment of the invention;
FIG. 2 is a schematic diagram of dual coupling of a hydrological model and an LSTM according to an embodiment of the present invention;
fig. 3 is a comparative graph of flow simulation results according to an embodiment of the present invention.
Detailed Description
The method and system for forecasting hydrologic under the influence of water taking activities based on double coupling of hydrologic model and LSTM are described in detail below with reference to the accompanying drawings.
Example 1
As shown in fig. 1, the method for forecasting hydrologic under the influence of water taking activities based on dual coupling of a hydrologic model and LSTM according to the present embodiment includes the following steps:
Step 1, constructing a distributed hydrological model according to the range of the forecast river basin, dividing the river basin and the river basin into grid units in the regional range, and dividing the sub-river basin based on the digital water system. And then, collecting historical data, and carrying out parameter calibration on the hydrologic model constructed in the forecast river basin to calibrate the model production convergence parameters.
The study area history data were: daily scale meteorological data comprising: rainfall, wind speed, temperature, relative humidity, and sunlight duration; drainage basin underlying data comprising: geographic elevation (DEM), land use type; the method for using the basic information of the water user comprises the following steps: taking a water taking source, a water type and a daily water taking amount monitoring value of a water user; hydrologic data, including soil humidity.
In the embodiment, the construction and parameter calibration work of the hydrologic model is specifically that hydrologic analysis is carried out on underlying surface data of a forecast river basin, parameter partitions and calculation units are divided, weights of different rainfall sites of each unit are deduced, and hydrologic model modeling is carried out. And according to the historical rainwater condition data, a multi-objective optimization algorithm is adopted to rate the yield and confluence parameters of each unit. The calibration objective function of the hydrologic model selects the Nash efficiency coefficient (NSE). Wherein, the calculation formula of NSE is:
Wherein: q sim,t and Q obs,t are the analog and actual values of the t-th time step, m 3/s, respectively; M 3/s, which is the average of the measured values; n represents the time series length. The maximum value of NSE is 1, and the closer the values of NSE and NSE are to 1, the better the simulation effect of the representative model.
And 2, according to the geographical positions of the water taking users, the water taking users are in one-to-one correspondence with grids and sub-watershed in the hydrologic model.
And step 3, performing space-time interpolation on the water-taking influence factors, and performing spatial interpolation on the stream domain attribute data to obtain all influence factor data corresponding to each grid.
The influencing factors are: daily scale meteorological data comprising: rainfall, wind speed, temperature, relative humidity, and sunlight duration; hydrologic data including soil humidity; socioeconomic performance, including: population density, total domestic production (GDP), night light index, irrigation area; drainage basin underlying data comprising: geographic elevation (DEM), land use type; the method for using the basic information of the water user comprises the following steps: taking water intake source, water type and daily water intake monitoring value of the water user.
And 4, after normalization processing is carried out on all the influence factor data and the daily water consumption, selecting an influence factor mean value in the influence area range of each water taking user as the input of the LSTM, taking the water consumption of the water taking user as a target value, and starting training the LSTM model.
The normalized calculation formula is:
Wherein: x i is a variable data in the sample set; x i max is the maximum value in the ith column of data in the sample set; x i min is the maximum value in the ith column of data in the sample set; x i' is the result of normalization of the sample set X i data; after the data has been normalized, the range of X i becomes [0,1].
In this embodiment, the LSTM model training parameters are set as follows: the training iteration number is 600, the number of hidden layers is 2, the number of hidden layer nodes is 32, the learning rate is 0.005, and the number of batch samples is 365. In this example, the root mean square error is used as the loss function, the tanh is used as the activation function, and the Adam algorithm is used as the optimizer.
And 5, coupling the LSTM model and the distributed hydrologic model. Based on the water resource condition and other influencing factors, predicting the water consumption through a trained network of the LSTM model; the predicted water usage amount is considered, the water resource amount is predicted based on a distributed hydrological model with a physical mechanism, and the predicted water resource amount becomes an influence factor of water usage amount prediction of the next time step. The rolling prediction is used for forming the mutual feedback of the water consumption and the water resource, and constructing a distributed hydrologic model under the influence of the human water consumption activity.
A. After the influence factor data of the day of prediction and the daily water consumption of the day three days before the day of prediction are subjected to normalization processing in the step S2, the daily water consumption of the day of prediction is predicted by using a trained LSTM network as the input of LSTM, and the result of the predicted water consumption is output to a distributed hydrological model;
B. Taking the influence of water taking activity into consideration in the production and collection process of the distributed hydrologic model, dividing a water taking mode into surface water taking and underground water taking, dividing the water taking mode into agricultural water and non-agricultural water, and identifying a water taking process for predicting an LSTM model;
C. Fig. 2 is a schematic diagram of model coupling. In the example, the distributed hydrologic model divides the watershed into a grid form, hydrologic simulation is carried out in the grid according to the principles of water balance and energy balance, and the influence of water taking activities on runoff is considered. The input of the distributed hydrologic model is meteorological data M such as daily precipitation P, daily temperature T (including daily average temperature, daily minimum temperature and daily maximum temperature), sunshine duration, relative humidity, surface 2M wind speed and the like, and river basin attribute related data B such as DEM, soil type, land utilization type and the like. The amount of stored water in the soil layer (soil moisture SM) and the subsurface aquifer (groundwater SG) is a model state variable. The daily runoff Q at the outlet of the river basin is the output of the distributed hydrological model. The input variables of the LSTM model include meteorological factors M, drainage basin attribute factors B, socioeconomic factors E, and water resource quantity factors (H). Meteorological factors, basin attribute factors and socioeconomic factors are taken as external input, basin outlet runoff Q and soil water content SM simulated by the distributed hydrologic model are taken as water resource quantity factor input (H), and are input into an LSTM network together, and the water taking quantity WW t+1 of a water taking user in the next time step is predicted based on a trained water taking prediction network.
The LSTM takes the meteorological data required for the water-rolling predictive model to be entered during operation, and the associated basin attribute data. In addition, the LSTM takes the initial value of the water consumption at time t 0 for the water rolling prediction model. During the operation process of the model, firstly, based on the water consumption at time t 0 and meteorological driving data of the current time step, simulating the hydrologic conditions in the river basin, wherein the hydrologic conditions comprise Q and SM:
Qt＝φ(B,Mt,WWt)
SMt＝g(B,Mt,SMt-1,WWt)
Wherein: phi and g respectively represent simulation processes of the improved WetSpa model on the runoff of the river basin and the water content of the soil under the driving of meteorological data and underlying surface data and the water data, and phi and g respectively correspond to the runoff of the river basin and the water content of the soil. Then, at the next time step, based on the trained LSTM water taking prediction network f (X), the hydrologic condition (H t) simulated by the previous day distributed hydrologic model, and the meteorological factor (M t+1) and the socioeconomic factor (E t+1) of the current time step, the water taking WW at time t+1 is predicted:
WWt+1＝f(B,Et,Mt,SMt,Qt)
Therefore, under the drive of meteorological data, underlying surface data and socioeconomic data, the coupling model carries out rolling simulation calculation on the water resource condition and the water consumption in the flow area, and realizes the prediction of the water consumption under the constraint of the water resource and the simulation of the runoff of the flow area under the influence of the water consumption of the society.
In the above process, there are two model couplings: the hydrologic model simulates the water resource amount, the water resource amount is used as an influence factor to predict the water taking amount, the water taking amount is used as one input of the hydrologic model, and then the water resource amount of the next time step can be simulated, so that the cycle iterates. And predicting an influence factor of the next time step (t+1) by adopting a distributed hydrological model, inputting the influence factor into an LSTM model, predicting the water taking amount of the next time step, taking the water taking activity into consideration in the hydrological model for hydrologic simulation based on the water taking amount, simulating the runoff of the step (t+2), and the like. The runoff prediction and water taking amount prediction under water taking activities can be realized under the condition of lacking hydrologic data and water taking data.
In this example, the method of the present invention described above was carried out by constructing a model in a Wei river basin. And setting 1 st of 2017 9 to 31 st of 2017 as a model preheating period, and predicting the outflow runoff of the river basin in 2018. As shown in fig. 3, the flow simulation results of the model (coupling model) constructed by the method of the present invention are more consistent with the actual flow than the most advanced prior art method (modified wetspa model). Based on the statistical index Nash efficiency coefficient (NSE) of the simulation value and the actual measurement value, the NSE of the coupling model simulation result is improved from 0.68 to 0.82 compared with the NSE of the original model simulation result, which shows that the method can improve the runoff simulation precision under the influence of the water taking activity of human beings.
In summary, the invention can realize the simulation of the runoff process under the influence of the water taking activity, and the water consumption simulation of the water shortage information area or the accurate and reliable prediction of the future water taking amount.
< Example two >
Further, the second embodiment provides a hydrological forecasting system based on the dual coupling of a hydrological model and an LSTM, which can automatically implement the method, and the system comprises a data set building part, a static network building part, a dynamic network building part, a model building part, a recognition part, an input display part and a control part.
The hydrologic model construction part can execute the content described in the step 1, construct a distributed hydrologic model according to the range of the forecast watershed, divide the sub-watershed and carry out parameter calibration.
The corresponding part of the water taking user can execute the content described in the step 2, and each water taking user corresponds to the grid unit and the sub-river basin in the hydrologic model one by one according to the geographic position of the water taking user.
The water usage impact factor obtaining section can execute the contents described in the above step 3 to obtain all water usage impact factor data corresponding to each grid.
The LSTM model training section can perform the contents described in step 4 above to train the LSTM model.
The coupling model building section can perform the above description of step 5, and the coupling LSTM model and the distributed hydrologic model build a coupling forecast model.
The input display part is used for enabling a user to input operation instructions and displaying input, output and intermediate processing data of the corresponding part in a text, table, graph, static or dynamic model mode according to the operation instructions.
The control part is communicated with the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part, the coupling model building part and the input display part, and controls the operation of the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part, the coupling model building part and the input display part.
The above embodiments are merely illustrative of the technical solutions of the present invention. The method and system for forecasting hydrologic under the influence of water taking activity based on double coupling of hydrologic model and LSTM are not limited to the above embodiments, but the scope of the invention is defined by the claims. Any modifications, additions or equivalent substitutions made by those skilled in the art based on this embodiment are within the scope of the invention as claimed in the claims.
Claims (10)
1. The hydrologic forecasting method under the influence of water taking activities based on the double coupling of a hydrologic model and an LSTM is characterized by comprising the following steps:
Step 1, constructing a distributed hydrological model according to the range of a forecast river basin, dividing the river basin into grid units, and dividing sub-river basins based on a digital water system; then, parameter calibration is carried out on the distributed hydrologic model according to the historical data;
step 2, according to the geographical positions of the water taking users, the water taking users are in one-to-one correspondence with grid cells and sub-watershed in the hydrologic model;
Step3, performing space-time interpolation on the water usage influence factors, and performing space interpolation on the stream domain attribute data to obtain all water usage influence factor data corresponding to each grid;
Step 4, after normalization processing is carried out on all the water taking influence factor data and the daily water taking quantity, selecting an influence factor mean value in the influence area range of each water taking user as an input of an LSTM model, taking the water taking quantity of the water taking user as a target value, and training the LSTM model;
step 5, coupling the LSTM model and the distributed hydrologic model to construct a coupling forecast model:
Based on the water taking influence factors, predicting the water taking amount through a trained LSTM model; then, inputting the predicted water consumption into a distributed hydrological model to predict the water resource quantity; the water resource quantity predicted by the distributed hydrologic model becomes a water taking influence factor of the LSTM model water taking quantity prediction of the next time step; the rolling prediction is used for forming the mutual coupling feedback of the water taking quantity and the water resource, constructing a coupling prediction model under the influence of the water taking activity of human beings, and performing simulation prediction on the water taking quantity and the hydrologic variable by the prediction coupling model.
2. The method for hydrologic forecasting under the influence of taking water activities based on hydrologic model and LSTM dual coupling according to claim 1, wherein:
In step 5, during the operation of the coupling prediction model, firstly, based on the water consumption at time t 0 and the meteorological driving data of the current time step, the hydrologic condition in the river basin is simulated, including Q and SM:
Qt＝φ(B,Mt,WWt)
SMt＝g(B,Mt,SMt-1,WWt)
wherein: phi and g respectively represent simulation processes of the model on runoff of a flow field and water content of soil under the driving of meteorological data, underlying surface data and water data;
Then, at the next time step, based on the already trained LSTM water taking prediction network f (X), the hydrologic condition H t simulated by the previous day distributed hydrologic model, and the meteorological factor M t+1, the socioeconomic factor E t+1 of the current time step, the water taking WW at time t+1 is predicted:
WWt+1＝f(B,Et,Mt,SMt,Qt)
Therefore, under the drive of meteorological data, underlying surface data and socioeconomic data, the coupling forecasting model carries out rolling simulation calculation on the water resource condition and the water taking quantity in the flow area, and the forecasting of the water taking quantity under the constraint of the water resource and the simulation of the runoff of the flow area under the influence of the water taking are realized.
3. The method for hydrologic forecasting under the influence of taking water activities based on hydrologic model and LSTM dual coupling according to claim 1, wherein:
In step 5, the influence factor data of the day of the prediction day and the daily water consumption of the day three before the prediction day are input into a trained LSTM model after normalization processing, the daily water consumption of the prediction day is predicted, and the result of the predicted water consumption is output to a distributed hydrological model;
Taking the influence of water taking activity into consideration in the production and collection process of the distributed hydrologic model, dividing a water taking mode into surface water taking and underground water taking, dividing the water taking mode into agricultural water and non-agricultural water, and identifying the water taking process predicted by the LSTM model;
The input variables of the LSTM model comprise meteorological factors, drainage basin attribute factors, socioeconomic factors and water resource quantity factors, the meteorological factors, the drainage basin attribute factors and the socioeconomic factors are used as external inputs, drainage basin outlet runoffs Q and soil water contents SM simulated by the distributed hydrologic model are used as water resource quantity factor inputs and are jointly input into the LSTM model, and the water taking quantity WW t+1 of a user taken in the next time step is predicted based on a trained water taking prediction network.
4. The method for hydrologic forecasting under the influence of taking water activities based on hydrologic model and LSTM dual coupling according to claim 1, wherein:
Wherein, in step3, taking the water influence factor includes: meteorological factors, socioeconomic factors, basin attribute factors, water history factors, and water resource quantity factors are taken.
5. The method for hydrologic forecasting under the influence of water taking activities based on hydrologic model and LSTM dual coupling according to claim 4, wherein:
the weather factors are daily scale weather data, and the weather factors comprise: rainfall, wind speed, temperature, relative humidity, and sunlight duration;
socioeconomic factors are socioeconomic developments, including: population density, domestic production total GDP, night light index, irrigation area;
the drainage basin attribute factor is drainage basin underlying data, comprising: geographic elevation DEM, land utilization type;
the taking water history factor is taking water user basic information, and comprises the following steps: taking a water taking source, a water type and a daily water taking amount monitoring value of a water user;
Water resource quantity factors, including runoff and soil moisture.
6. Hydrologic forecasting system under taking water activity influence based on hydrologic model and LSTM dual coupling, characterized in that includes:
A hydrological model construction unit which constructs a distributed hydrological model according to the range of the forecast basin, divides the basin into grid cells, and divides the sub-basin based on the digital water system; then, parameter calibration is carried out on the distributed hydrologic model according to the historical data;
The water taking user corresponding part is used for carrying out one-to-one correspondence on each water taking user and grid cells and sub-watershed in the hydrologic model according to the geographic position of the water taking user;
A water usage influence factor obtaining part for performing space-time interpolation on the water usage influence factors and performing spatial interpolation on the stream domain attribute data to obtain all water usage influence factor data corresponding to each grid;
The LSTM model training part is used for carrying out normalization processing on all the water taking influence factor data and the daily water taking quantity, selecting an influence factor mean value in the influence area range of each water taking user as the input of the LSTM model, taking the water taking quantity of the water taking user as a target value, and training the LSTM model;
A coupling model construction part for constructing a coupling forecast model by coupling the LSTM model and the distributed hydrologic model: based on the water taking influence factors, predicting the water taking amount through a trained LSTM model; then, inputting the predicted water consumption into a distributed hydrological model to predict the water resource quantity; the water resource quantity predicted by the distributed hydrologic model becomes a water taking influence factor of the LSTM model water taking quantity prediction of the next time step; the rolling prediction is used for forming the mutual coupling feedback of the water taking quantity and the water resource, constructing a coupling prediction model under the influence of the water taking activity of human beings, and performing simulation prediction on the water taking quantity and the hydrologic variable by the prediction coupling model;
The control part is communicated with the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part and the coupling model building part, and controls the operation of the hydrologic model building part, the water taking user corresponding part, the water taking influence factor obtaining part, the LSTM model training part and the coupling model building part.
7. The hydrologic forecasting system under the influence of water activity based on dual coupling of hydrologic model and LSTM of claim 6, further comprising:
And the input display part is in communication connection with the control part, allows a user to input an operation instruction, and displays input, output and intermediate processing data of the corresponding part in a text, form, graph, static or dynamic model mode according to the operation instruction.
8. The hydrologic forecasting system under the influence of water usage activity based on hydrologic model and LSTM dual coupling of claim 6, wherein:
in the operation process of the coupling forecasting model, firstly, based on the water consumption at time t 0 and meteorological driving data of the current time step, simulating the hydrologic condition in a river basin, wherein the coupling forecasting model comprises Q and SM:
Qt＝φ(B,Mt,WWt)
SMt＝g(B,Mt,SMt-1,WWt)
wherein: phi and g respectively represent simulation processes of the model on runoff of a flow field and water content of soil under the driving of meteorological data, underlying surface data and water data;
Then, at the next time step, based on the already trained LSTM water taking prediction network f (X), the hydrologic condition H t simulated by the previous day distributed hydrologic model, and the meteorological factor M t+1, the socioeconomic factor E t+1 of the current time step, the water taking WW at time t+1 is predicted:
WWt+1＝f(B,Et,Mt,SMt,Qt)
Therefore, under the drive of meteorological data, underlying surface data and socioeconomic data, the coupling forecasting model carries out rolling simulation calculation on the water resource condition and the water taking quantity in the flow area, and the forecasting of the water taking quantity under the constraint of the water resource and the simulation of the runoff of the flow area under the influence of the water taking are realized.
9. The hydrologic forecasting system under the influence of water usage activity based on hydrologic model and LSTM dual coupling of claim 6, wherein:
in the coupling prediction model, the influence factor data of the day of the prediction day and the daily water consumption of the day three days before the prediction day are input into a trained LSTM model after normalization processing, the daily water consumption of the prediction day is predicted, and a result of the predicted water consumption is output to a distributed hydrological model;
Taking the influence of water taking activity into consideration in the production and collection process of the distributed hydrologic model, dividing a water taking mode into surface water taking and underground water taking, dividing the water taking mode into agricultural water and non-agricultural water, and identifying the water taking process predicted by the LSTM model;
The input variables of the LSTM model comprise meteorological factors, drainage basin attribute factors, socioeconomic factors and water resource quantity factors, the meteorological factors, the drainage basin attribute factors and the socioeconomic factors are used as external inputs, drainage basin outlet runoffs Q and soil water contents SM simulated by the distributed hydrologic model are used as water resource quantity factor inputs and are jointly input into the LSTM model, and the water taking quantity WW t+1 of a user taken in the next time step is predicted based on a trained water taking prediction network.
10. The hydrologic forecasting system under the influence of water usage activity based on hydrologic model and LSTM dual coupling of claim 6, wherein:
Wherein taking the water influence factor includes: meteorological factors, socioeconomic factors, basin attribute factors, water history factors and water resource quantity factors are taken;
The meteorological factors are day scale meteorological data comprising: rainfall, wind speed, temperature, relative humidity, and sunlight duration;
socioeconomic factors are socioeconomic developments, including: population density, domestic production total GDP, night light index, irrigation area;
the drainage basin attribute factor is drainage basin underlying data, comprising: geographic elevation DEM, land utilization type;
the taking water history factor is taking water user basic information, and comprises the following steps: taking a water taking source, a water type and a daily water taking amount monitoring value of a water user;
Water resource quantity factors, including runoff and soil moisture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311518460.9A CN118095034B (en) | 2023-11-15 | 2023-11-15 | Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311518460.9A CN118095034B (en) | 2023-11-15 | 2023-11-15 | Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118095034A true CN118095034A (en) | 2024-05-28 |
| CN118095034B CN118095034B (en) | 2024-09-17 |
Family
ID=91146692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311518460.9A Active CN118095034B (en) | 2023-11-15 | 2023-11-15 | Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118095034B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118350512A (en) * | 2024-06-12 | 2024-07-16 | 湖南华博信息技术有限公司 | Data analysis method of intelligent water quantity prediction model based on ultra-short-term data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577895A (en) * | 2013-11-08 | 2014-02-12 | 河海大学 | Method for forecasting monthly runoff through secondary coupling under condition of data shortage |
| CN114386334A (en) * | 2022-01-19 | 2022-04-22 | 浙江大学 | Runoff rolling forecasting method based on distributed hydrological runoff simulation substitution model |
| WO2022110582A1 (en) * | 2020-11-26 | 2022-06-02 | 浙江大学 | Runoff forecasting method based on baseflow separation and artificial neural network model |
| CN115115148A (en) * | 2022-08-30 | 2022-09-27 | 武汉大学 | Long-term runoff forecasting method and system based on process-data collaborative driving |
| CN115238513A (en) * | 2022-08-02 | 2022-10-25 | 北京师范大学 | River basin runoff ensemble forecasting method considering climate and land utilization changes |
| CN116643331A (en) * | 2023-05-17 | 2023-08-25 | 昆明思永科技有限公司 | Hydrologic forecasting method based on hydrologic information big data of regional river basin |
-
2023
- 2023-11-15 CN CN202311518460.9A patent/CN118095034B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577895A (en) * | 2013-11-08 | 2014-02-12 | 河海大学 | Method for forecasting monthly runoff through secondary coupling under condition of data shortage |
| WO2022110582A1 (en) * | 2020-11-26 | 2022-06-02 | 浙江大学 | Runoff forecasting method based on baseflow separation and artificial neural network model |
| CN114386334A (en) * | 2022-01-19 | 2022-04-22 | 浙江大学 | Runoff rolling forecasting method based on distributed hydrological runoff simulation substitution model |
| CN115238513A (en) * | 2022-08-02 | 2022-10-25 | 北京师范大学 | River basin runoff ensemble forecasting method considering climate and land utilization changes |
| CN115115148A (en) * | 2022-08-30 | 2022-09-27 | 武汉大学 | Long-term runoff forecasting method and system based on process-data collaborative driving |
| CN116643331A (en) * | 2023-05-17 | 2023-08-25 | 昆明思永科技有限公司 | Hydrologic forecasting method based on hydrologic information big data of regional river basin |
Non-Patent Citations (4)
| Title |
|---|
| CHEN S: "Improving daily streamflow simulations for data-scarce watersheds using the coupled SWAT-LSTM approach", 《JOURNAL OF HYDROLOGY》, 31 July 2023 (2023-07-31) * |
| 刘双: "基于神经网络与半分布式水文模型相结合的缺资料区径流估计模型——以莺落峡流域为例", 《北京师范大学学报(自然科学版)》, 15 June 2016 (2016-06-15) * |
| 吴梦琪: "分布式水文模型EasyDHM在增江流域的应用研究", 《中国农村水利水电》, 15 November 2019 (2019-11-15) * |
| 郝春沣;贾仰文;王浩;: "气象水文模型耦合研究及其在渭河流域的应用", 水利学报, no. 09, 15 September 2012 (2012-09-15) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118350512A (en) * | 2024-06-12 | 2024-07-16 | 湖南华博信息技术有限公司 | Data analysis method of intelligent water quantity prediction model based on ultra-short-term data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118095034B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113379109B (en) | Runoff forecasting method based on prediction model self-adaption | |
| CN115507822B (en) | Flood risk prediction method driven by hydrologic cycle variation | |
| CN102567635B (en) | Method for quantificationally distinguishing contributions of different factors in water cycle evolution process | |
| CN112765800A (en) | Design method of distributed water resource configuration model | |
| CN107292098A (en) | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology | |
| CN110363356A (en) | A kind of water and soil resources space and time optimization configuration method of ecology-oriented | |
| CN118095034B (en) | Hydrologic forecasting method and system based on hydrologic model and LSTM dual coupling under influence of water taking activity | |
| CN104484715A (en) | Neural network and particle swarm optimization algorithm-based building energy consumption predicting method | |
| CN111539597B (en) | Gridding drainage basin social and economic drought assessment method | |
| Wang et al. | Integrated simulation of the dualistic water cycle and its associated processes in the Haihe River Basin | |
| CN108614915B (en) | Hydrological model free construction strategy method based on scene driving | |
| CN116485584B (en) | Method and system for cooperative regulation and control of WEE of river with large bottom slope in alpine region | |
| CN102495937A (en) | Prediction method based on time sequence | |
| CN111598724B (en) | Time-interval integration method for day-ahead prediction of warehousing flow of small and medium reservoirs | |
| CN113723703A (en) | Water quality prediction method and system based on multi-source data fusion and deep learning | |
| CN112052559A (en) | Three-parameter hydrological model based on monthly water balance and construction method thereof | |
| CN116562583A (en) | Multidimensional water resource supply and demand prediction method and system | |
| CN112700068A (en) | Reservoir dispatching rule optimization method based on machine learning fusion of multi-source remote sensing data | |
| Song et al. | Framework on low-carbon retrofit of rural residential buildings in arid areas of northwest China: A case study of Turpan residential buildings | |
| CN117610208A (en) | River water quality and quantity prediction method based on urban water system model interaction | |
| CN115935667A (en) | Regional surface water resource prediction simulation and optimal configuration method thereof | |
| CN111428936A (en) | River basin rainfall flood availability index measuring and calculating method based on distributed water nodes | |
| CN117172037B (en) | Distributed hydrologic forecasting method, device, computer equipment and medium | |
| CN117455041B (en) | Multi-target management and control utilization method for future water resources obtained through prediction | |
| CN117236552A (en) | Regional water demand prediction method based on multi-level index system |
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 |