CN113642699A - Intelligent river flood forecasting system - Google Patents
Intelligent river flood forecasting system Download PDFInfo
- Publication number
- CN113642699A CN113642699A CN202110736926.7A CN202110736926A CN113642699A CN 113642699 A CN113642699 A CN 113642699A CN 202110736926 A CN202110736926 A CN 202110736926A CN 113642699 A CN113642699 A CN 113642699A
- Authority
- CN
- China
- Prior art keywords
- flood
- forecasting
- river
- data
- neural network
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C13/00—Surveying specially adapted to open water, e.g. sea, lake, river or canal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Abstract
The invention discloses an intelligent river flood forecasting system, which comprises a flood forecasting model (10), a database (20), a water regime forecasting subsystem (30) and a client (40); wherein: the flood forecasting model (10) comprises a single river channel flood forecasting model and a river system flood forecasting model; for different flood forecasting modes, learning the evolution law of river channel flood by training by utilizing a neural network, and forecasting the downstream flood process according to the upstream flood process; establishing a single river channel and river system flood forecasting model by adopting an intelligent algorithm combining an improved BP neural network and a genetic algorithm for forecasting river channel flood; and the user reads the relevant data information in the database by calling the flood forecasting model to forecast the flood. Compared with the prior art, the method adopts the method of combining the improved BP neural network and the genetic algorithm which introduce the peak correction coefficient to forecast, has reasonable modeling, reliable algorithm, higher precision and greatly improved running speed compared with the prior art.
Description
Technical Field
The invention relates to the field of emergency disaster prevention and a computer application technology, in particular to an intelligent river flood forecasting model system.
Background
The ice flood monitoring technology is an effective water conservancy information technology and has important practical significance for guaranteeing life and property safety of people, but at present, ice flood monitoring still stays in methods of hydrologic section observation, field patrol and the like, artificial observation has many limitations, ice condition information collection is difficult to meet the demand of ice prevention decision making, and river and terrain information of the whole area cannot be reflected globally and intuitively to support the decision maker to make a reasonable ice prevention and flood prevention scheduling decision. The ice flood influence factors are many and complex, and various ice flood prevention measures still need to be comprehensively applied at the present stage to relieve or eliminate the ice flood disasters.
With the rapid development of modern information technology, water conservancy informatization drives the progress of water conservancy modernization, and digital water conservancy technology provides an important agenda for water conservancy development. Modern high and new technologies such as a geographic information system (GIS-geographic information system), remote sensing (RS-RemoteSensing), graphic processing, network communication (Internet), DataBase (DB-DataBase) management, virtual reality (VR-virtual reality) and the like are applied to water conservancy informatization construction and are a necessary trend of water conservancy development. In recent years, remote sensing technology is rapidly developed, high-resolution image data are easier to obtain, and large-range ground observation information can be provided due to the fact that the remote sensing technology has the characteristics of real-time performance, continuity and the like. With the improvement of various monitoring means, various monitoring information is more comprehensive and needs standardized management, the flood prevention department can timely obtain the freezing sealing condition of a large-range river channel by using the system, the dangerous situations which possibly occur in the river sealing and opening periods are found, the occurred dangerous situations are reasonably evaluated, and effective protection measures are taken.
Disclosure of Invention
Aiming at the problems that the ice flood monitoring means is backward, and river and terrain information of the whole area cannot be reflected globally and intuitively to support reasonable flood prevention and ice prevention scheduling decisions of decision makers, the invention adopts the technologies of remote sensing, data processing, virtual reality and the like to realize real-time monitoring, global effective management and display of river ice flood information and provide basis for ice prevention emergency disaster reduction.
Aiming at the problems in the prior art, the invention aims to develop an intelligent river flood forecasting model system, improve timeliness and accuracy of flood forecasting, reduce flood risks and alleviate flood disasters.
The invention has the significance of improving the monitoring technical means, monitoring the ice slush occurrence and development process in real time by using the prior art, and effectively making decisions to reduce casualties and reduce disaster loss.
Compared with the prior art, the invention can achieve the following beneficial effects:
the invention adopts the method of introducing the peak correction coefficient and combining the improved BP neural network and the genetic algorithm to forecast, has reasonable modeling, reliable algorithm and higher precision, and greatly improves the running speed compared with the prior art.
Drawings
Fig. 1 is a schematic diagram of an architecture of an intelligent river flood forecasting system according to the present invention;
FIG. 2 is a schematic diagram of a calculation process of the water regime forecasting subsystem;
FIG. 3 is a schematic diagram of a single river flood forecasting model;
fig. 4 is a schematic diagram of an intelligent forecasting model of river flood, wherein: t represents a previous time, and T represents a forecast period;
fig. 5 is a schematic diagram of an embodiment of an intelligent river flood forecasting system;
FIG. 6 is a diagram illustrating the comparison between the predicted results of the present invention and the prior art.
Detailed Description
The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
Fig. 1 is a schematic diagram of an intelligent river flood forecasting system according to the present invention. The system mainly comprises a flood forecasting model 10, a database 20, a water situation forecasting subsystem 30 and a client 40. Wherein:
the flood forecasting model 10 comprises a single river channel flood forecasting model and a river system flood forecasting model; single river flood forecasting means that flood flows in a main river, and no other flood flows in or out of other large branches; river flood forecasting refers to forecasting flood evolution of a river system consisting of a main stream river channel and one or more branch stream river channels; for different flood forecasting modes, the neural network learns the evolution law of river flood through training, so that the downstream flood process is forecasted through association according to the upstream flood process and the network. The flood forecasting model 10 adopts an intelligent algorithm combining an improved BP neural network and a genetic algorithm to establish a single river channel and river system flood forecasting model for forecasting river channel flood; and the user can read the related data information in the database by calling the flood forecasting model to forecast the flood.
The database 20 is used for storing data;
the water regime forecasting subsystem 30 is used for providing inquiry and analysis of basin basic information and water and rain regime information in the interaction between the water regime forecasting subsystem data and the user;
and the client 40 is used for providing a visual query interface of a prediction result for a user, facilitating the operation of the user and reading data. The request and the response between the client 40 and the river flood intelligent forecasting system are realized through flood forecasting operation 50.
The intelligent river flood forecasting system is divided into a system presentation layer, a system application layer, a system service layer, a data access layer and a data layer from top to bottom on an implementation level. Data collection, arrangement and system design scheme determination, namely, planning each application system to be built according to user requirements, and dividing system boundaries; analyzing the function and operation requirement of each subsystem, and planning a required support and service platform; and determining the required software and hardware systems and network communication indexes of the computer according to the requirements of the support and service platform.
Fig. 2 is a schematic flow chart of the flood forecasting model 10. The flood forecasting model 10 comprises a genetic algorithm part and a BP algorithm part, and specifically comprises the following steps:
the genetic algorithm part comprises:
extracting flood data including flow and water level in a data base as input data; preprocessing flood data, specifically normalizing the flood data; initializing a weight population (including the number of hidden layers, a learning rate and a peak correction coefficient, wherein the value of the peak correction coefficient is related to the maximum value and the minimum value of a training sample, the general value range is 1.5-2.5, initializing the evolution times, population scale, cross probability and variation probability, carrying out real number coding on the population, and taking the error between predicted data and expected data as a fitness function); constructing a genetic operator for optimizing the neural weight and the threshold value of the river flood forecast BP within the value space of the initial weight and the threshold value of the flood forecast BP neural network, and generating offspring by utilizing a crossover operator; outside the value space of the initial weight and the threshold of the flood forecasting BP neural network, generating a second generation and a third generation by using a mutation operator to perform genetic calculation; (ii) a Circularly executing the series of operations of selection, intersection and variation until the number of evolutionary times is reached to obtain the optimal initialization weight and threshold; constructing a BP neural network by using the obtained optimal initial weight and a threshold;
the BP algorithm portion includes the following processes: training a BP neural network by using training data, and determining a topological structure of the BP neural network; determining a calculation parameter; obtaining BP neural network weight; testing the neural network by using the test data in combination with the input data after the preprocessing, and performing inverse normalization processing on the predicted data; analyzing the error between the predicted data and the expected data, and calculating the error of the BP neural network; modifying the network weight according to a network error introduced peak value correction theory; and outputting a BP algorithm result until the training times are finished.
The specific method for modifying the network weight comprises the following steps:
the weights of the network are modified in the direction of decreasing the peak training error by introducing reasonable correction coefficients for the error of the maximum. The vector expression of the error after adding the correction coefficient is as follows:
in the formula: NL is the total number of neurons,
refers to the maximum of all the outputs,
is the correction coefficient of the training error, f is the transfer function, and here, a logarithmic unipolar Sigmoid function is adopted.
For hidden neurons, the error vector expression when the training number is k is as follows:
the corresponding weight vector expression is:
the invention realizes the research and development of the river system intelligent forecasting model. Flood forecasting modes for river channels basically include a single river channel flood forecasting mode and a river system flood forecasting mode. Selecting the improved BP neural network model as a modeling basis, taking water level (flow) data of an upstream main flow station and a main branch flow station as the input of the network model, and taking a corresponding water level (flow) formed by a downstream flow station as the output of the network model, so that the network can map the corresponding water level (flow); meanwhile, the propagation time of the flood from the upstream hydrological station to the downstream hydrological station is the forecast period of the network for the flood.
Fig. 3 is a schematic diagram of a single river flood forecasting mode. The single river flood intelligent forecasting is the simplest forecasting form, and a network model of the single river flood intelligent forecasting is shown as (3a) in fig. 3. The flood forecast model for a longer river can be regarded as a series form of the sub-models, as shown in (3b) of fig. 3.
Fig. 4 is a schematic diagram of an intelligent river flood forecasting model. Compared with a single river channel model, the flood process of three continuous periods of main flow and each branch flow is used as input and output of the network, and the following flood evolution information is given to the network:
(1) the network simulates the continuous process of flooding;
(2) in the input item, the upstream main flow and each branch flow simulate the amplitude of each time interval in the upstream flood process through the water levels (flow rates) of three continuous time intervals; in the output item, the water level (flow) of the downstream main flow in three continuous periods simulates the amplitude of the downstream flood in each period;
(3) in the input item, the downstream main flow simultaneous water level (flow) is used as the input of a network model to simulate the influence of the downstream initial water level (flow).
Fig. 5 is a schematic diagram of an intelligent river flood forecasting system according to an embodiment. The specific embodiment is a trunk and branch river system flood forecasting model system of the west river section of the Yangtze river basin, and the neural network forecasting system is formed by the neural network submodels intersected by the trunk and branch in a series-parallel connection mode. The time interval of flood data acquisition is 3h, the time flow of an upstream main flow river transfer station at T moment, the time flow of a tributary nanning station (T-8), the time flow of a pair pavilion station (T-3), the time flow of a Liuzhou station (T-3) and the time flow of a downstream main flow estuary station at T moment are used as the input of a network, the time flow of the estuary station (T + T) is the output of the network, wherein T is the average meeting period, and the average value of flood propagation time among stations is taken.
Fig. 6 is a schematic diagram comparing the predicted results of the present invention with the prior art.
Compared with the prior art, the technical scheme of the invention has the following beneficial effects:
the method adopts a method of combining an improved BP neural network and a genetic algorithm which introduce a peak correction coefficient to forecast, has reasonable modeling, reliable algorithm and higher precision, and greatly improves the running speed compared with the prior art, wherein the peak correction is a mode based on deep learning, and the peak prediction precision is effectively improved; in addition, the intelligent flood forecasting system compiled by adopting VC + + language optimizes the model calculation method, realizes the man-machine interaction of the system, increases the dynamic graphic display function of the operation result, is convenient and intuitive to operate, and is easier to popularize and apply. The development and development of the system exert the advantage that the neural network can map a complex system, so that flood control decision can be better served.
The present invention is not limited to the above-mentioned preferred embodiments, and any other products in various forms can be obtained by anyone in the light of the present invention, but any changes in the shape or structure thereof, which have the same or similar technical solutions as the present application, fall within the protection scope of the present invention.
Claims (2)
1. An intelligent river flood forecasting system is characterized by comprising a flood forecasting model (10), a database (20), a water condition forecasting subsystem (30) and a client (40); wherein:
the flood forecasting model (10) comprises a single river channel flood forecasting model and a river system flood forecasting model; for different flood forecasting modes, learning the evolution law of river channel flood by training by utilizing a neural network, and forecasting the downstream flood process according to the upstream flood process; establishing a single river channel and river system flood forecasting model by adopting an intelligent algorithm combining an improved BP neural network and a genetic algorithm for forecasting river channel flood; a user reads related data information in a database by calling a flood forecasting model to forecast the flood;
the database (20) is used for data storage;
the water condition forecasting subsystem (30) is used for providing inquiry and analysis of basin basic information and water and rain condition information in the interaction of the water condition forecasting subsystem data and a user;
and the client (40) is used for providing a query interface with a visual prediction result for the user, supporting the user operation and reading data.
2. The intelligent river flood forecasting system according to claim 1, wherein the flood forecasting model (10) comprises a genetic algorithm part and a BP algorithm part, and the following are specific:
the genetic algorithm part comprises:
extracting flood data including flow and water level in a data base as input data; preprocessing flood data, specifically normalizing the flood data; initializing a weight population; constructing a genetic operator for optimizing the neural weight and the threshold value of the river flood forecast BP within the value space of the initial weight and the threshold value of the flood forecast BP neural network, and generating offspring by utilizing a crossover operator; outside the value space of the initial weight and the threshold of the flood forecasting BP neural network, generating a second generation and a third generation by using a mutation operator to perform genetic calculation; circularly executing the series of operations of selection, intersection and variation until the number of evolutionary times is reached to obtain the optimal initialization weight and threshold; constructing a BP neural network by using the obtained optimal initial weight and a threshold;
the BP algorithm part comprises:
training a BP neural network by using training data, and determining a topological structure of the BP neural network; determining a calculation parameter; obtaining BP neural network weight; testing the neural network by using the test data in combination with the input data after the preprocessing, and performing inverse normalization processing on the predicted data; analyzing the error between the predicted data and the expected data, and calculating the error of the BP neural network; modifying the network weight according to a network error introduced peak value correction theory; and outputting a BP algorithm result until the training times are finished.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110736926.7A CN113642699A (en) | 2021-06-30 | 2021-06-30 | Intelligent river flood forecasting system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110736926.7A CN113642699A (en) | 2021-06-30 | 2021-06-30 | Intelligent river flood forecasting system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113642699A true CN113642699A (en) | 2021-11-12 |
Family
ID=78416399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110736926.7A Pending CN113642699A (en) | 2021-06-30 | 2021-06-30 | Intelligent river flood forecasting system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113642699A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115374376A (en) * | 2022-10-24 | 2022-11-22 | 水利部交通运输部国家能源局南京水利科学研究院 | Small hydropower station ecological influence monitoring and evaluating method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110471950A (en) * | 2019-07-19 | 2019-11-19 | 河海大学 | A kind of middle and small river Real-time Flood Forecasting model of mind forecasting procedure |
| CN111199298A (en) * | 2018-11-19 | 2020-05-26 | 贺州市水利局 | Flood forecasting method and system based on neural network |
| CN111597758A (en) * | 2020-05-14 | 2020-08-28 | 河海大学 | Medium and small river integrated forecasting method based on negative correlation learning |
| AU2020102997A4 (en) * | 2020-10-24 | 2020-12-24 | Neeraj Kumar | A system and a process for flood extent mapping and identification of flood risk zones |
-
2021
- 2021-06-30 CN CN202110736926.7A patent/CN113642699A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199298A (en) * | 2018-11-19 | 2020-05-26 | 贺州市水利局 | Flood forecasting method and system based on neural network |
| CN110471950A (en) * | 2019-07-19 | 2019-11-19 | 河海大学 | A kind of middle and small river Real-time Flood Forecasting model of mind forecasting procedure |
| CN111597758A (en) * | 2020-05-14 | 2020-08-28 | 河海大学 | Medium and small river integrated forecasting method based on negative correlation learning |
| AU2020102997A4 (en) * | 2020-10-24 | 2020-12-24 | Neeraj Kumar | A system and a process for flood extent mapping and identification of flood risk zones |
Non-Patent Citations (1)
| Title |
|---|
| 苑希民: "人工神经网络与遗传算法在河道洪水预报中的应用", 《水利发展研究》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115374376A (en) * | 2022-10-24 | 2022-11-22 | 水利部交通运输部国家能源局南京水利科学研究院 | Small hydropower station ecological influence monitoring and evaluating method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492822B (en) | Air pollutant concentration time-space domain correlation prediction method | |
| CN107610464B (en) | A kind of trajectory predictions method based on Gaussian Mixture time series models | |
| CN109142171B (en) | Urban PM10 concentration prediction method based on feature expansion and fusing with neural network | |
| CN109508360B (en) | Geographical multivariate stream data space-time autocorrelation analysis method based on cellular automaton | |
| CN106650767B (en) | Flood forecasting method based on cluster analysis and real-time correction | |
| CN107730054B (en) | Gas load combined prediction method based on support vector regression | |
| Yu et al. | Prediction of highway tunnel pavement performance based on digital twin and multiple time series stacking | |
| CN113537600B (en) | Medium-long-term precipitation prediction modeling method for whole-process coupling machine learning | |
| CN107610021A (en) | The comprehensive analysis method of environmental variance spatial and temporal distributions | |
| CN104408900A (en) | Dynamic optimization based neural network flood warning device and method | |
| Huangpeng et al. | Forecast of the hydropower generation under influence of climate change based on RCPs and Developed Crow Search Optimization Algorithm | |
| CN110555551B (en) | Air quality big data management method and system for smart city | |
| CN112735097A (en) | Regional landslide early warning method and system | |
| CN110363349A (en) | A kind of LSTM neural network hydrologic(al) prognosis method and system based on ASCS | |
| CN113033110B (en) | Important area personnel emergency evacuation system and method based on traffic flow model | |
| CN117236199B (en) | Method and system for improving water quality and guaranteeing water safety of river and lake in urban water network area | |
| CN113554466A (en) | Short-term power consumption prediction model construction method, prediction method and device | |
| CN111968019A (en) | Landslide hazard susceptibility prediction method based on CNN and RNN integration | |
| CN109299208A (en) | Transmission tower intelligent visual methods of risk assessment under a kind of typhoon disaster | |
| CN111199298A (en) | Flood forecasting method and system based on neural network | |
| CN114693064A (en) | Building group scheme generation performance evaluation method | |
| CN114492233B (en) | Watershed water simulation method based on webGIS platform and considering comprehensive utilization requirements | |
| Wu et al. | Local and global Bayesian network based model for flood prediction | |
| CN113642699A (en) | Intelligent river flood forecasting system | |
| CN109190800A (en) | A kind of sea surface temperature prediction technique based on spark frame |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20211112 |
|
| WD01 | Invention patent application deemed withdrawn after publication |