CN113609783B - Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information - Google Patents
Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information Download PDFInfo
- Publication number
- CN113609783B CN113609783B CN202110944776.9A CN202110944776A CN113609783B CN 113609783 B CN113609783 B CN 113609783B CN 202110944776 A CN202110944776 A CN 202110944776A CN 113609783 B CN113609783 B CN 113609783B
- Authority
- CN
- China
- Prior art keywords
- upward
- forecasting
- salt tide
- tide
- model
- 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
- 150000003839 salts Chemical class 0.000 title claims abstract description 221
- 238000000034 method Methods 0.000 title abstract description 27
- ZAMOUSCENKQFHK-UHFFFAOYSA-N Chlorine atom Chemical compound [Cl] ZAMOUSCENKQFHK-UHFFFAOYSA-N 0.000 claims abstract description 21
- 229910052801 chlorine Inorganic materials 0.000 claims abstract description 21
- 239000000460 chlorine Substances 0.000 claims abstract description 21
- 238000007637 random forest analysis Methods 0.000 claims description 37
- 238000012795 verification Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 25
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 238000012216 screening Methods 0.000 claims description 15
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000007405 data analysis Methods 0.000 claims description 6
- 230000010355 oscillation Effects 0.000 claims description 5
- 238000013277 forecasting method Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 5
- 238000003066 decision tree Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000011144 upstream manufacturing Methods 0.000 description 4
- 230000007774 longterm Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Images
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information. The large-scale circulating climate information coupled salt tide upward-tracing forecasting system acquires hydrological meteorological data which historically influences the upward tracing of the salt tide from the area needing to be forecasted, screens out key influence factors from the data to establish a salt tide upward-tracing forecasting model, verifies the salt tide upward-tracing forecasting model, and verifies the chlorine content output by the passing salt tide upward-tracing forecasting model as the forecasting data of the upward tracing of the salt tide in a certain area; the forecast data obtained by the salt tide upward-tracing forecasting system coupled with the large-scale circulating climate information is high in precision, and the technical problems that the existing salt tide upward-tracing forecasting model is high in data requirement, high in information acquisition cost and low in forecasting precision due to the fact that forecasting factors are mutually interfered are solved.
Description
Technical Field
The invention relates to the technical field of environmental data models, in particular to a salt tide upward forecasting system and method coupled with large-scale circulation climate information.
Background
The upward tracing of salt tide is taken as a typical water environment problem in estuary regions, and the problems occur in coastal or coastal countries or regions. The serious upward backtracking of the salt tide can not only change the water quality condition in the riverway and destroy the water ecological balance, but also disturb the water supply order of surrounding cities and restrict the local social and economic development. Therefore, the key influence factors and the time lag of the salt tide upward are researched, and a medium-long-term forecasting mechanism of the salt tide upward is established, so that the method has important practical significance on regional water resource management.
The existing methods for forecasting the salt tide upward of the estuary comprise three methods, namely an empirical formula method, a physical experiment method and a numerical simulation method. Wherein: the empirical formula method is mainly to establish an empirical formula between the salt tide and the influencing elements thereof and deduce the mathematical relationship between the salt tide and the influencing elements based on the statistics and analysis of a large amount of measured data. For example, davie prandtl in estuary dynamics and morphology summarizes the chlorinity simulation equation applicable to a partially mixed estuary, where the magnitude of chlorinity at a point is related to the cross-sectional area of the estuary, the attenuation coefficient, the longitudinal mixing coefficient, the runoff rate, and the distance from the point to the estuary, but this empirical formula ignores the time factor of the horizontal convective diffusion equation. The physical experiment method is mainly a physical model or a field test method, and is used for reducing and simulating a physical process. For example, by field testing at teweiter estuary, uk, which analyzed the response of the chlorinity to runoff and tide, it was found that the richardson number of the estuary reached the empirical maximum at the time of ebb tide and the minimum at the time of flood tide. A physical model test of the upward tracing of the salt tide of the knife sharpening gate water channel is developed in the Ruchen and what use of the salt tide intrusion and salt suppression technology of the knife sharpening gate, and the upward tracing change characteristics of the salt tide of the knife sharpening gate water channel are observed by changing the influence factors of upstream incoming water, sea level, tidal range and the like. The numerical simulation method is mainly to establish a mathematical model and solve the mathematical model by a computer method. For example, an unstructured finite-volume offshore ocean three-dimensional model FVCOM is adopted to establish a numerical model for tracing from a sharpening gate water channel to a large and small ocean salt tide, and the space-time change rule and the vertical layering phenomenon of the chlorine content of the sharpening gate water channel are researched and analyzed.
However, most of the three methods have high requirements on application conditions or data and are complex to operate, so that the research on the upstream forecasting of the salt tide in the area with data shortage is limited. And, the main factors influencing the upward tracing of the salt tide include upstream incoming water, tide, sea level, wind, estuary topography and the like. At present, the three methods for the upward-tracing of the salt tide do not consider the application of large-scale circulation climate factors to the upward-tracing medium-long term forecast of the salt tide, and a salt tide upward-tracing forecast model is mostly constructed through daily-scale runoff and tidal change, so that the simulation research on the chlorine content of months and other time scales is generally less, and the upward-tracing precision of the simulated salt tide is low.
Disclosure of Invention
The embodiment of the invention provides a salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information, which are used for solving the technical problems that the existing salt tide upward-tracing forecasting model has high requirement on data and high information acquisition cost, and has the interference of correlated forecasting factors, so that the forecasting precision is low.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
a salt tide upward-tracing forecasting system coupled with large-scale circulating climate information comprises a data input module, a model forecasting module and a data output module;
the data input module is used for acquiring the hydrological meteorological data influencing the salt tide up-tracking, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the salt tide up-tracking;
the model forecasting module is used for establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
and the data output module is used for outputting the optimal model for the salt tide upward forecast to forecast the chlorine content of the area affected by the salt tide upward.
Preferably, the data input module comprises a data acquisition sub-module, a data analysis sub-module and a screening sub-module;
the data acquisition submodule is used for acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain the hydrological meteorological data of the salt tide up-tracking;
the data analysis submodule is used for respectively carrying out random importance analysis on the hydrological meteorological data influencing the salt tide up-tracking through the estimation error and the purity of the sample set outside the bag to obtain a first importance number set and a second importance number set between each factor and the chlorinity under different time scales;
the screening submodule is used for respectively screening the factors in the first importance degree set and the second importance degree from large to small according to the importance degrees to obtain a corresponding first factor set and a corresponding second factor set; and selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking.
Preferably, the hydrologic factors include flow and tide level, and the large-scale circulating climate factor includes pacific interpersonal oscillation and southern billow index.
Preferably, the screening submodule is configured to screen out factors with importance ranks in the first two thirds to construct a first factor set and a second factor set, respectively, after sorting the factors in the first importance set and the second importance set from high to low.
Preferably, the model forecasting module comprises a number set classification submodule, a model building submodule and a model verification submodule;
the number set classification submodule is used for constructing a number set by all the key influence factors and dividing the number set into a training set and a verification set;
the model building submodule is used for carrying out model training on the training set by adopting a random forest algorithm and building a salt tide upward forecasting model;
and the verification model submodule is used for inputting the verification set into the saltwater uptracking forecasting model, outputting a Nash efficiency coefficient and a decision coefficient corresponding to the verification set, and obtaining a saltwater uptracking forecasting optimal model if the Nash efficiency coefficient and the decision coefficient are both greater than a coefficient threshold value.
Preferably, the number set classification submodule is configured to use 80% of the number set as a training set and 20% of the number set as a verification set.
Preferably, the coefficient threshold is 0.5.
The invention also provides a salt tide upward forecasting method coupled with the large-scale circulation climate information, which comprises the following steps:
acquiring hydrological meteorological data influencing the backtracking of the salt tide, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the backtracking of the salt tide;
establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
and outputting the optimal model for forecasting the upward tracking of the salt tide to forecast the chlorine content of the area affected by the upward tracking of the salt tide.
Preferably, the step of obtaining the key influence factor influencing the salt tide up-tracking through analyzing the acquired hydrometeorological data by the random forest importance degree comprises the following steps:
acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain hydrological meteorological data of salt tide up-tracking:
respectively carrying out random importance analysis on the hydrological meteorological data influencing the salt tide up-tracking through estimation errors and the purity of the sample set outside the bag to obtain a first importance set and a second importance set between each factor and the chlorinity under different time scales;
sorting all factors in the first importance degree set and the second importance degree according to the importance degrees, and then screening to obtain a corresponding first factor set and a second factor set; and selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking.
Preferably, the step of obtaining the optimal model for the salt tide up-tracking forecast comprises the following steps:
constructing a number set of all the obtained key influence factors, and dividing the number set into a training set and a verification set;
performing model training on the training set by adopting a random forest algorithm, and establishing a salt tide upward forecasting model;
and inputting the verification set into the saltwater uptracking forecasting model, outputting a Nash efficiency coefficient and a decision coefficient corresponding to the verification set, and obtaining a saltwater uptracking forecasting optimal model if the Nash efficiency coefficient and the decision coefficient are both larger than a coefficient threshold value.
According to the technical scheme, the embodiment of the invention has the following advantages: the system and the method for forecasting the upward saltwater tide coupled with the large-scale circulating climate information comprise a data input module, a model forecasting module and a data output module, wherein the data input module is used for acquiring hydrological meteorological data influencing the upward saltwater tide, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the upward saltwater tide; the model forecasting module is used for establishing a salt tide upward forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward forecasting model to obtain an optimal salt tide upward forecasting model; and the data output module is used for outputting the optimal model for the salt tide upward forecast to forecast the chlorine content of the area affected by the salt tide upward. The large-scale circulating climate information coupled salt tide upward-tracing forecasting system acquires hydrological meteorological data which historically influences the upward tracing of the salt tide from the area needing to be forecasted, screens out key influence factors from the data to establish a salt tide upward-tracing forecasting model, verifies the salt tide upward-tracing forecasting model, and verifies the chlorine content output by the passing salt tide upward-tracing forecasting model as the forecasting data of the upward tracing of the salt tide in a certain area; the forecast data obtained by the salt tide upward-tracing forecasting system coupled with the large-scale circulating climate information is high in precision, and the technical problems that the existing salt tide upward-tracing forecasting model is high in data requirement, high in information acquisition cost and low in forecasting precision due to the fact that forecasting factors are mutually interfered are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a frame diagram of a salt tide upward forecasting system coupled with large-scale circulating climate information according to an embodiment of the present invention.
Fig. 2 is a random forest algorithm frame diagram adopted by the salt tide upward forecasting system coupled with large-scale circulating climate information according to the embodiment of the invention.
Fig. 3 is a frame diagram of a data input module of the salt tide upward forecasting system coupled with large-scale circulating climate information according to the embodiment of the present invention.
Fig. 4 shows key influence factors of the salt tide upward forecasting system coupled with large-scale circulating climate information according to the embodiment of the present invention.
Fig. 5 is a frame diagram of a salt tide upward forecasting system model forecasting module coupled with large-scale circulating climate information according to an embodiment of the present invention.
Fig. 6 is a monthly forecasting flow chart of the salt tide upward forecasting system model forecasting module coupled with the large-scale circulating climate information according to the embodiment of the present invention.
Fig. 7 is a forecast monthly scale salt tide forecast map based on hydrologic influence factors by the data output module of the salt tide upward-tracing forecast system coupled with large-scale circulating climate information according to the embodiment of the present invention.
Fig. 8 is a monthly scale salt tide forecasting graph forecast by the salt tide upward-tracking forecasting system data output module coupled with the large-scale circulating climate information based on the combination of hydrology and large-scale circulating climate influence factors according to the embodiment of the invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the application provides a salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information, and the system and method are used for solving the technical problems that an existing salt tide upward-tracing forecasting model is high in data requirement, high in information acquisition cost and low in forecasting precision due to the fact that forecasting factors are mutually interfered.
The first embodiment is as follows:
fig. 1 is a frame diagram of a salt tide upward forecasting system coupled with large-scale circulating climate information according to an embodiment of the present invention, and fig. 2 is a frame diagram of a random forest algorithm adopted by the salt tide upward forecasting system coupled with large-scale circulating climate information according to an embodiment of the present invention.
As shown in fig. 1 and fig. 2, an embodiment of the present invention provides a salt tide upward forecasting system coupled with large-scale circumfluence climate information, which includes a data input module 10, a model forecasting module 20, and a data output module 30;
the data input module 10 is used for acquiring the hydrographic meteorological data influencing the salt tide up-tracking, and analyzing the acquired hydrographic meteorological data through random forest importance to obtain a key influence factor influencing the salt tide up-tracking;
the model forecasting module 20 is used for establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
and the data output module 30 is used for outputting the salt tide upward forecast optimal model to forecast the chlorine content of the area affected by the salt tide upward.
In the embodiment of the present invention, the data input module 10 mainly obtains the hydrographic meteorological data affecting the salt tide up-tracking in a certain area since the history, and screens out the key influence factors affecting the salt tide up-tracking from the obtained hydrographic meteorological data by using the random forest importance analysis.
In the embodiment of the present invention, the model forecasting module 20 mainly establishes a salt tide upward-tracing forecasting model by using a random forest algorithm for the screened key influence factors, then verifies the obtained salt tide upward-tracing forecasting model, and uses the verified salt tide upward-tracing forecasting model as an optimal salt tide upward-tracing forecasting model.
In the embodiment of the present invention, the data output module 30 mainly outputs the chlorine content output by the obtained optimal model for forecasting the upward movement of the salt tide, and the chlorine content output by the optimal model for forecasting the upward movement of the salt tide is forecast data for forecasting the upward movement of the salt tide in a certain area.
The invention provides a salt tide upward-tracing forecasting system coupled with large-scale circulating climate information, which comprises a data input module, a model forecasting module and a data output module, wherein the data input module is used for acquiring hydrological meteorological data influencing upward tracing of salt tides and analyzing the acquired hydrological meteorological data through random forest importance to obtain key influence factors influencing upward tracing of the salt tides; the model forecasting module is used for establishing a salt tide upward forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward forecasting model to obtain an optimal salt tide upward forecasting model; and the data output module is used for outputting the optimal model for the salt tide upward forecast to forecast the chlorine content of the area affected by the salt tide upward. The large-scale circulating climate information coupled salt tide upward-tracing forecasting system acquires hydrological meteorological data which historically influences the upward tracing of the salt tide from the area needing to be forecasted, screens out key influence factors from the data to establish a salt tide upward-tracing forecasting model, verifies the salt tide upward-tracing forecasting model, and verifies the chlorine content output by the passing salt tide upward-tracing forecasting model as the forecasting data of the upward tracing of the salt tide in a certain area; the forecast data obtained by the salt tide upward-tracing forecasting system coupled with the large-scale circulating climate information is high in precision, and the technical problems that the existing salt tide upward-tracing forecasting model is high in data requirement, high in information acquisition cost and low in forecasting precision due to the fact that forecasting factors are mutually interfered are solved.
Fig. 3 is a frame diagram of a data input module of the salt tide upward forecasting system coupled with large-scale circulating climate information according to the embodiment of the present invention, and fig. 4 is a key influence factor of the salt tide upward forecasting system coupled with large-scale circulating climate information according to the embodiment of the present invention.
As shown in fig. 3 and 4, in an embodiment of the present invention, the data input module 10 includes a data obtaining sub-module 11, a data analyzing sub-module 12 and a screening sub-module 13;
the data acquisition submodule 11 is used for acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain the hydrological meteorological data of the salt tide up-tracking;
the data analysis submodule 12 is used for respectively carrying out random importance analysis on the hydrological meteorological data influencing the salt tide up-tracking through the estimation error and the purity of the sample set outside the bag to obtain a first importance number set and a second importance number set between each factor and the chlorinity under different time scales;
the screening submodule 13 is configured to sort the factors in the first importance degree and the second importance degree according to the importance degrees, and then screen the factors to obtain a corresponding first factor set and a corresponding second factor set; selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking;
the hydrological factors comprise flow, tide level and the like, and the large-scale circular current climate factors comprise pacific annual oscillation, southern billow index and the like.
It should be noted that the screening submodule 13 is mainly configured to screen out factors with importance levels ranked in the first two thirds to construct the first factor set and the second factor set correspondingly after sorting the factors in the first importance level set and the second importance level set from high to low respectively. The data input module 10 comprehensively and meticulously discusses the remote correlation and time lag between the 14 large-scale circulating climate factors and the upward-tracing activity of the salt tide in a certain area based on the evaluation of the importance of the random forest, and determines the combination of forecast factors of the upward tracing of the salt tide at different time scales.
In the embodiment of the invention, the data input module 10 mainly uses the data such as flow, tidal level and the like, the pacific annual oscillation, southern surge index and the like, which are acquired by the data acquisition submodule 11, to form the hydrological meteorological data which influence the salt tide up-tracking; then, through the analysis of the random forest importance of the data analysis submodule 12, the correlation between each factor and the chlorinity under different time scales of day, month, year and the like in the data is explored, namely the importance is obtained; finally, the screening submodule 13 screens out the factors with the same importance degree ranked at the top 2/3 from the first importance degree set and the second importance degree set as key influence factors influencing the backtracking of the salt tide.
It should be noted that the higher the importance degree is, the more closely the relationship between the expression factor and the chlorinity is, so that the key influence factor which influences the strength of the upstream activity of the salt tide in the estuary region of a certain area can be obtained. The obtained key influence factors are used as input data of the salt tide upward forecasting model, so that the accuracy of the forecasting data output by the obtained optimal salt tide upward forecasting model is high. The optimal model for forecasting the upward movement of the salt tide considers the correlation between the large-scale circulation climate factor and the upward movement of the salt tide, so that the capture and extraction of the upward movement characteristics of the salt tide by the optimal model for forecasting the upward movement of the salt tide are enhanced, and the obtained optimal model for forecasting the upward movement of the salt tide can be effectively applied to the medium-term and long-term forecasting of the salt tide in the area with data shortage due to the easy acquireability of the input data of the model.
In the embodiment of the invention, the importance evaluation of the estimation error of the sample set outside the bag is also called displacement importance evaluation, the importance of the variable is evaluated mainly by comparing the change of the model effect before and after the random displacement interpretation variable value, and the larger the change is, the more important the interpretation variable plays in the model simulation process is. The calculation formula is as follows:
in the formula, VIM (x)i)OOBAn importance index for the ith explanatory variable evaluated from the out-of-bag sample set estimation error; n is the number of decision trees; OOBerrjAnd OOBerr'jAnd respectively estimating errors for the out-of-bag sample sets of the jth decision tree before and after the replacement. The explanatory variables refer to respective influence factors for evaluating the importance.
In embodiments of the present invention, the degree of kini impurity refers to the frequency at which sub-items are randomly drawn from a data set, the sub-items being mislabeled. For randomly drawn sub-items X1,X2,...,XmThe degree of purity (i.e., the frequency with which a sub-item is incorrectly labeled) of the kini is:
in the formula, fiAs random samples XiProbability of being correctly marked.
In the formula, IGt(f) And IGr(f) Are respectively asAfter the node is split, the corresponding kiney purities of the two new nodes are obtained; n is the number of decision trees; k is an explanatory variable xiThe number of occurrences of the m node in the jth decision tree.
Fig. 5 is a frame diagram of a module for forecasting a salt tide upward-looking forecasting system model coupled with large-scale circulating climate information according to an embodiment of the present invention, and fig. 6 is a flowchart of a month forecasting process of the module for forecasting a salt tide upward-looking forecasting system coupled with large-scale circulating climate information according to an embodiment of the present invention.
As shown in FIG. 5, in one embodiment of the present invention, the model forecasting module 20 includes an number set classification submodule 21, a model building submodule 22, and a model verification submodule 23;
the number set classification submodule 21 is used for constructing a number set by all the obtained key influence factors and dividing the number set into a training set and a verification set;
the model building submodule 22 is used for carrying out model training on the training set by adopting a random forest algorithm and building a salt tide upward forecasting model;
and the verification model submodule 23 is configured to input the verification set into the salt tide upward-tracing forecasting model, output a nash efficiency coefficient and a decision coefficient corresponding to the verification set, and obtain an optimal salt tide upward-tracing forecasting model if both the nash efficiency coefficient and the decision coefficient are greater than a coefficient threshold value.
Note that the coefficient threshold is preferably 0.5. And the number set classification submodule is used for taking 80% of the salt tide upward-tracing data and the number set as a training set and taking 20% of the salt tide upward-tracing data and the number set as a verification set. The model forecasting module 20 adopts a random forest algorithm to establish a salt tide upward forecasting model coupled with large-scale circulation climate information.
As shown in fig. 6, in the embodiment of the present invention, the model prediction module 20 adopts a random forest algorithm to construct a salt tide upward-tracing prediction model of a estuary region in a certain place under different time scales, gradually inputs key influence factors in a training set into the salt tide upward-tracing prediction model based on the random forest algorithm according to importance, combines existing chlorinity data to train and verify the salt tide upward-tracing prediction model, and finally selects an input variable combination with an optimal fitting degreeAnd the forecasting factor is used as a forecasting factor of the optimal model for forecasting the upward forecasting of the salt tide to carry out upward forecasting of the salt tide. In this embodiment, taking the forecast of the month scale of the salt tide upward as an example, the establishment and application of the salt tide upward forecast model are respectively as follows: the salt tide upward forecasting model sets different input variable scenes, wherein: the hydrologic factor is input in a scene 1, the hydrologic coupling large-scale circulation climate factor is input in a scene 2, and the large-scale circulation climate factor is input in a scene 3. Inputting the three situations into a salt tide upward-tracing forecasting model respectively, and enabling the salt tide upward-tracing forecasting model to achieve the optimal fitting effect under the three situations through training and verification, wherein the fitting effect of the model adopts a Nash efficiency coefficient (NSE) and a decision coefficient (R)2) Evaluation was performed. And applying the trained optimal model for the upward forecasting of the salt tide to the upward forecasting of the salt tide, and comparing the forecasting precision under three conditions.
Fig. 7 is a monthly-scale salt tide forecast map forecasted based on a hydrologic influence factor by the salt tide upward-tracing forecast system data output module coupled with large-scale circulating climate information according to the embodiment of the present invention, and fig. 8 is a monthly-scale salt tide forecast map forecasted based on the combination of hydrologic and large-scale circulating climate influence factor by the salt tide upward-tracing forecast system data output module coupled with large-scale circulating climate information according to the embodiment of the present invention.
In the embodiment of the invention, if monthly scale prediction is taken as an example, the salt tide upward-tracing forecasting system coupled with the large-scale circulating climate information selects data of hydrologic influence factors such as flow rate and tide level of the sharpening gate water channel in 2005 + 2015 year and large-scale circulating factors such as pacific annual oscillation and southern wave motion index, and researches the correlation degree between the salt tide upward tracing of the sharpening gate water channel under different time scales and the influence factors thereof through random forest importance analysis, so that the key influence factor influencing the salt tide upward tracing of the region is identified. And training and verifying the salt tide upward forecasting model by adopting salt tide upward data and number sets, wherein 80% of the salt tide upward data and number sets are used as a training set of the salt tide upward forecasting model, and 20% of the salt tide upward data and number sets are used as a verification set of the salt tide upward forecasting model. As shown in fig. 7 and 8, in the verification process, the fitting effect of the salt tide upward forecasting model forecasted based on the combination of hydrology and large-scale circulating climate influence factors is optimal, and the obtained NSE and R2 values are 0.80 and 0.82 respectively, so that the forecasting effect of the model can be obviously improved by coupling large-scale circulating climate information, and the method has an important practical significance for guiding the salt tide upward forecasting of the area lacking data.
The salt tide upward-tracing forecasting system coupled with the large-scale circulation climate information can bring the large-scale circulation climate factor into salt tide upward-tracing medium and long-term forecasting, and shows that the large-scale circulation climate factor under the influence of remote correlation has feasibility and effectiveness in the long-term forecasting of the salt tide; in order to effectively avoid redundancy of forecast information and cross correlation among forecast factors, the salt tide upward-tracing forecasting system determines forecast elements and time delay of salt tide upward tracing under different time scales through a comparison and selection method, and the salt tide upward-tracing forecasting system is low in requirement on data information and simple and convenient to operate.
Example two:
the invention also provides a salt tide upward forecasting method coupled with the large-scale circulation climate information, which comprises the following steps:
acquiring hydrological meteorological data influencing the backtracking of the salt tide, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the backtracking of the salt tide;
establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
and outputting a salt tide upward forecast optimal model to forecast the chlorine content of the area affected by the salt tide upward.
In the embodiment of the invention, the step of analyzing the acquired hydrographic meteorological data through random forest importance to obtain the key influence factors influencing the salt tide up-tracking comprises the following steps:
acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain hydrological meteorological data of salt tide up-tracking:
respectively carrying out random importance analysis on hydrological meteorological data influencing the salt tide up-tracking through estimation errors and the purity of the sample set outside the bag to obtain a first importance set and a second importance set between each factor and the chlorinity under different time scales;
sorting the factors in the first importance degree set and the second importance degree according to the importance degrees from large to small, and then screening to obtain a corresponding first factor set and a corresponding second factor set; and selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking.
In the embodiment of the invention, the step of obtaining the optimal model for the salt tide upward-tracing forecasting comprises the following steps:
constructing a number set of all the obtained key influence factors, and dividing the number set into a training set and a verification set;
performing model training on the training set by adopting a random forest algorithm, and establishing a salt tide upward forecasting model;
inputting the verification set into the salt tide upward-tracing forecasting model, outputting a Nash efficiency coefficient and a decision coefficient corresponding to the verification set, and if the Nash efficiency coefficient and the decision coefficient are both larger than a coefficient threshold value, obtaining the optimal salt tide upward-tracing forecasting model.
It should be noted that the steps in the second method in the embodiment correspond to the module settings in the first system in the embodiment, and the contents of each module have been described in detail in the first system in the embodiment, and the contents of the steps in the second method in the embodiment are not further described in detail again.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (6)
1. A salt tide upward-tracing forecasting system coupled with large-scale circulating climate information is characterized by comprising a data input module, a model forecasting module and a data output module;
the data input module is used for acquiring the hydrological meteorological data influencing the salt tide up-tracking, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the salt tide up-tracking;
the model forecasting module is used for establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
the data output module is used for outputting the optimal model for the salt tide upward forecast to forecast the chlorine content of the area affected by the salt tide upward forecast;
the data input module comprises a data acquisition sub-module, a data analysis sub-module and a screening sub-module;
the data acquisition submodule is used for acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain the hydrological meteorological data of the salt tide up-tracking;
the data analysis submodule is used for respectively carrying out random forest importance analysis on the hydrometeorological data influencing the salt tide up-tracking through the estimation error and the purity of the sample set outside the bag to obtain a first importance number set and a second importance number set between each factor and the chlorinity under different time scales;
the screening submodule is used for respectively screening the factors in the first importance degree set and the second importance degree from large to small according to the importance degrees to obtain a corresponding first factor set and a corresponding second factor set; selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking;
the model forecasting module comprises a number set classification submodule, a model building submodule and a model verifying submodule;
the number set classification submodule is used for constructing a number set by all the key influence factors and dividing the number set into a training set and a verification set;
the model building submodule is used for carrying out model training on the training set by adopting a random forest algorithm and building a salt tide upward forecasting model;
and the verification model submodule is used for inputting the verification set into the saltwater uptracking forecasting model, outputting a Nash efficiency coefficient and a decision coefficient corresponding to the verification set, and obtaining a saltwater uptracking forecasting optimal model if the Nash efficiency coefficient and the decision coefficient are both greater than a coefficient threshold value.
2. The salt tide up-forecast system coupled with large-scale circulating climate information according to claim 1, wherein said hydrologic factors comprise flow and tide level, and said large-scale circulating climate factors comprise pacific annual oscillation and southern surge index.
3. The system for backtracking and forecasting of salt tide coupled with large-scale circulating climate information according to claim 1, wherein the screening submodule is configured to screen out factors with importance degree ranked in the first two thirds after sorting the factors in the first importance degree set and the second importance degree from high to low respectively to construct a first factor set and a second factor set.
4. The system of claim 1, wherein the number set classification submodule is configured to use 80% of the number set as a training set and 20% of the number set as a validation set.
5. The system of claim 1, wherein the coefficient threshold is 0.5.
6. A salt tide upward-tracing forecasting method coupled with large-scale circulation climate information is characterized by comprising the following steps:
acquiring hydrological meteorological data influencing the backtracking of the salt tide, and analyzing the acquired hydrological meteorological data through random forest importance to obtain a key influence factor influencing the backtracking of the salt tide;
establishing a salt tide upward-tracing forecasting model by adopting a random forest algorithm according to the key influence factors, and verifying the salt tide upward-tracing forecasting model to obtain an optimal salt tide upward-tracing forecasting model;
outputting the optimal model for forecasting the upward tracking of the salt tide to forecast the chlorine content of the area affected by the upward tracking of the salt tide;
the step of analyzing the acquired hydrographic meteorological data through random forest importance to obtain a key influence factor influencing the salt tide up-tracking comprises the following steps:
acquiring historical hydrological factors, large-scale circulation climate factors and chlorine content of estuary areas of a certain place to obtain hydrological meteorological data of salt tide up-tracking:
respectively carrying out random forest importance analysis on the hydrological meteorological data influencing the salt tide up-tracking through estimation errors and the purity of the sample set outside the bag to obtain a first importance number set and a second importance number set between each factor and the chlorinity under different time scales;
sorting all factors in the first importance degree set and the second importance degree according to the importance degrees, and then screening to obtain a corresponding first factor set and a second factor set; selecting the same factor from the first factor set and the second factor set as a key influence factor influencing the salt tide up-tracking;
the step of obtaining the optimal model for the salt tide up-tracking forecast comprises the following steps:
constructing a number set of all the obtained key influence factors, and dividing the number set into a training set and a verification set;
performing model training on the training set by adopting a random forest algorithm, and establishing a salt tide upward forecasting model;
and inputting the verification set into the saltwater uptracking forecasting model, outputting a Nash efficiency coefficient and a decision coefficient corresponding to the verification set, and obtaining a saltwater uptracking forecasting optimal model if the Nash efficiency coefficient and the decision coefficient are both larger than a coefficient threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944776.9A CN113609783B (en) | 2021-08-17 | 2021-08-17 | Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944776.9A CN113609783B (en) | 2021-08-17 | 2021-08-17 | Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113609783A CN113609783A (en) | 2021-11-05 |
| CN113609783B true CN113609783B (en) | 2022-05-03 |
Family
ID=78341010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110944776.9A Active CN113609783B (en) | 2021-08-17 | 2021-08-17 | Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609783B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117709242B (en) * | 2024-02-05 | 2024-04-09 | 珠江水利委员会珠江水利科学研究院 | Estuary salty tide tracing strength evaluation method |
| CN117709603B (en) * | 2024-02-06 | 2024-05-17 | 珠江水利委员会珠江水利科学研究院 | Estuary salty tide early warning grade classification method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992960A (en) * | 2017-11-03 | 2018-05-04 | 中山大学 | Hekou District salty tide traces back chlorosity Simulation prediction method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9579700B2 (en) * | 2014-05-30 | 2017-02-28 | Iteris, Inc. | Measurement and modeling of salinity contamination of soil and soil-water systems from oil and gas production activities |
-
2021
- 2021-08-17 CN CN202110944776.9A patent/CN113609783B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992960A (en) * | 2017-11-03 | 2018-05-04 | 中山大学 | Hekou District salty tide traces back chlorosity Simulation prediction method |
Non-Patent Citations (4)
| Title |
|---|
| Modeling saltwater intrusion using an integrated Bayesian model averaging method in the Pearl River Delta;Lin, Kairong 等;《JOURNAL OF HYDROINFORMATICS》;20191101;第21卷(第6期);全文 * |
| 基于Delft3D-flow模型的长江口盐度扩散规律模拟;李林娟等;《人民长江》;20161214(第23期);全文 * |
| 基于随机森林模型的咸潮预报;苏程佳等;《热带地理》;20180502(第03期);全文 * |
| 闽江咸潮上溯与保障供水安全综合措施的现状与思考;陈奔月;《水利科技》;20200301(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113609783A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111310968B (en) | LSTM neural network circulating hydrologic forecasting method based on mutual information | |
| Li et al. | Investigating a complex lake-catchment-river system using artificial neural networks: Poyang Lake (China) | |
| Davison et al. | Statistical modeling of spatial extremes | |
| Mendoza et al. | An intercomparison of approaches for improving operational seasonal streamflow forecasts | |
| Tencaliec et al. | Flexible semiparametric generalized Pareto modeling of the entire range of rainfall amount | |
| CN113609783B (en) | Salt tide upward-tracing forecasting system and method coupled with large-scale circulating climate information | |
| CN111027175A (en) | Method for evaluating social and economic influences of flood based on coupling model integrated simulation | |
| Niroomandi et al. | Extreme value analysis of wave climate in Chesapeake Bay | |
| Gilleland et al. | Spatial extreme value analysis to project extremes of large‐scale indicators for severe weather | |
| Apurv et al. | Evaluation of the stationarity assumption for meteorological drought risk estimation at the multidecadal scale in contiguous United States | |
| Loukas et al. | Evaluation of climate change on drought impulses in Thessaly, Greece | |
| CN112016839A (en) | Flood disaster prediction and early warning method based on QR-BC-ELM | |
| Chakraborty et al. | Impact of extreme events on conversion efficiency of wave energy converter | |
| Kim et al. | Probabilistic coastal storm surge analyses using synthesized tracks based on historical typhoon parameters | |
| Vincent | Mesoscale wind fluctuations over Danish waters | |
| Mu et al. | The NAO variability prediction and forecasting with multiple time scales driven by ENSO using machine learning approaches | |
| Yan et al. | A stochastic tropical cyclone model for the northwestern Pacific Ocean with improved track and intensity representations | |
| Beaver et al. | Pattern-based evaluation of coupled meteorological and air quality models | |
| CN109870146A (en) | Flood frequency calculation method and device under a kind of climate change | |
| Zhao et al. | Spatial–temporal distribution and forecasting model of precipitation using dynamic-statistical information fusion | |
| Denamiel et al. | Next-generation (sub-) kilometre-scale climate modelling for extreme storm-surge hazard projections | |
| CN113094876B (en) | Reservoir flood control risk calculation method and system based on ARIMA model under future situation | |
| Smithers et al. | Development and assessment of a continuous simulation modelling system for design flood estimation | |
| Smit | Extreme Waves in the North Sea: Deriving extreme wave conditions applying Hierarchical Clustering and Non-Stationary Extreme Value Modelling | |
| Sohrabi et al. | Evaluation of a stochastic weather generator for long-term ensemble streamflow forecasts |
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 | ||
| OL01 | Intention to license declared |