CN113139700A - River flow prediction method, device, equipment and storage medium - Google Patents
River flow prediction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113139700A CN113139700A CN202110545815.8A CN202110545815A CN113139700A CN 113139700 A CN113139700 A CN 113139700A CN 202110545815 A CN202110545815 A CN 202110545815A CN 113139700 A CN113139700 A CN 113139700A
- Authority
- CN
- China
- Prior art keywords
- information
- hydrological station
- historical
- upstream
- station
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- 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
- G06N3/045—Combinations of networks
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a river flow prediction method, a device, equipment and a storage medium, wherein the method comprises the following steps: acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; acquiring appointed historical rainfall information within a preset range from the target hydrological station; acquiring appointed historical flow information of a target hydrological station; calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station; acquiring related information of an upstream hydrological station and related information of a tributary hydrological station; acquiring rainfall information in the preset range; the future flow information of the target hydrological station is predicted according to the related information of the upstream hydrological station, the related information of the branch hydrological station, the rainfall information and the time lag information, the problem that the river flow prediction method in the prior art is low in accuracy is solved, and the accuracy of the river flow prediction method is improved.
Description
Technical Field
The invention relates to the technical field of river flow prediction, in particular to a river flow prediction method, a device, equipment and a storage medium.
Background
Forecasting and forecasting hydrologic information are always the key points of concern in natural environment management in various countries. China also continuously adds hydrological stations with automatic monitoring equipment in recent years, including hydrological stations, water level stations and rainfall stations, to perfect collection of hydrological conditions of rivers, lakes and reservoirs. In the application aspect of the water regime data, besides the first monitoring data is used for analysis and evaluation, the real-time monitoring data is used for predicting the future water regime change condition and trend, so that scientific support is provided for decisions such as water use scheduling, flood and drought management, hydraulic engineering implementation and the like. In forecasting the water regime, one of the core targets is flow forecasting.
At present, flow prediction mainly has two modes, namely, simulation of target watershed hydrodynamic force is realized through a mechanism model, boundary condition setting and parameter calibration, and then the prediction purpose is achieved through input of predicted simulation boundary conditions. More applied models such as river hydraulic analysis (HEC-RAS); the other is to predict the future by a statistical model through fitting the correlation between data and/or self-variation trend and rule. Such as pearson type iii curves.
The various index factors in the water environment field mostly have a direct or indirect mechanistic relationship, for example, rainfall influences the river flow in the form of surface runoff. However, due to the complexity of natural environment, the relationship is difficult to be completely generalized by a biochemical mechanism at present, and instead, the rules and the relationship between numerical values can be summarized in a black box by using a machine learning method, so that the method is further applied to the requirements of prediction and the like. With the wide construction of automatic stations in recent years, the stability of data stream transmission, the improvement of data quality, the improvement of monitoring indexes and the increase of data magnitude all provide good soil for flow prediction of a deep learning algorithm based on a statistical model. Some network models requiring large data, such as convolutional networks, Long Short-Term Memory (LSTM) and the like, have been successfully applied to traffic prediction.
With the wide application of the machine learning model in various industries, the machine learning model also has some achievements in the flow prediction of the water environment. Predicting river flow based on a BP neural network model, for example, according to the multi-section water level; or predicting river flow based on a convolutional network according to flow-related characteristic factors, and the like. The former considers that similar change rules (relevance) exist in the monitored water level and the flow rate between the upstream and the downstream of the river, a group of independent variables are formed by water level data of adjacent upstream and upstream hydrological stations measured at the same time, and a network model is formed by matching and fitting the flow rate at the corresponding time through a BP neural network. And then the river flow is predicted through a water level-flow relation curve. The latter screens the factor composition characteristics strongly related to the flow data, and then generates a time sequence data sample training model for flow prediction.
The above two methods have searched for river flow prediction from the perspective of the upstream-downstream relationship and the multivariate correlation on the spatial level, but have the following problems:
1. which ignores the delay problem of factors over time. For example, in 2020, the water level in the upper reaches of the Yangtze river is continuously increased under the influence of strong rainfall in the near days of about 6 and 20 months, and the water level of the Sanxia dam exceeds the flood control warning line to 6 and 21 months. At this time, for the hydrological stations at the downstream of the Yangtze river, the influence of the upstream flow can be transmitted after a period of time delay due to factors such as the flow transmission distance, the flow limitation of a series of gate dams on the riverway and the like. For flow prediction, one such scenario, which combines spatial propagation delay and artifacts, is difficult to accurately model with a mechanism model; and secondly, the correlation between the upstream monitoring data and the downstream monitoring data is difficult to predict through the two network models.
2. Only the upstream and downstream monitoring sections are considered, but the flow of the downstream monitoring section is influenced by the upstream and the branch in many ways. For example, in a red river in a Guizhou province, the tributary contributes half of the total monitoring flow rate downstream of the river, so that the monitoring data upstream and downstream of the river have only weak correlation and cannot be used for prediction simulation.
Therefore, how to design an efficient and accurate river flow prediction method becomes a key point for the technical problem to be solved and the research of the technical staff in the field.
Disclosure of Invention
In view of this, embodiments of the present invention provide a river discharge prediction method, apparatus, device, and storage medium, so as to solve the problem in the prior art that the river discharge prediction method is low in accuracy.
Therefore, the embodiment of the invention provides the following technical scheme:
in a first aspect of the present invention, a river discharge prediction method is provided, which includes the following steps:
acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; the designated historical related information of the upstream hydrological station comprises designated historical flow information of the upstream hydrological station and/or designated historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station;
acquiring appointed historical rainfall information within a preset range from the target hydrological station;
acquiring appointed historical flow information of the target hydrological station;
calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station;
acquiring relevant information of the upstream hydrological station and relevant information of the tributary hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the tributary hydrological station comprises flow information of the tributary hydrological station and/or water level information of the tributary hydrological station;
acquiring rainfall information in the preset range;
and predicting future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
Optionally, calculating time lag information according to the specified history related information of the upstream hydrological station, the specified history related information of the tributary hydrological station, the specified historical rainfall information, and the specified historical flow information of the target hydrological station includes:
recording the lag value of a time node t corresponding to the appointed historical flow information of the target hydrological station as 0;
let the lag value of time node t-n be 2n(ii) a Wherein n represents an integer between 1 and | logm |; m is the maximum allowable delay amount;
selecting the designated historical related information of the upstream hydrological station with the highest degree of correlation with the designated historical flow information of the target hydrological station from the historical related information of a plurality of upstream hydrological stations corresponding to the time node t-n;
acquiring a lag value corresponding to the appointed historical relevant information of the upstream hydrological station;
calculating first time lag information according to the lag value corresponding to the appointed historical relevant information of the upstream hydrological station;
selecting the appointed historical related information of the tributary hydrological station with the highest degree of correlation with the appointed historical flow information of the target hydrological station from the historical related information of a plurality of tributary hydrological stations corresponding to the time node t-n;
acquiring a lag value corresponding to the appointed historical relevant information of the tributary hydrological station;
calculating second time lag information according to the lag value corresponding to the appointed historical relevant information of the tributary hydrological station;
selecting the appointed historical rainfall information with the highest correlation degree with the appointed historical flow information of the target hydrological station from a plurality of historical rainfall information corresponding to the time node t-n;
acquiring a lag value corresponding to the designated historical rainfall information;
calculating third time lag information according to the lag value corresponding to the designated historical rainfall information;
and obtaining the time lag information according to the first time lag information, the second time lag information and the third time lag information.
Optionally, the method further comprises:
acquiring historical related information of an upstream hydrological station, historical related information of a tributary hydrological station and historical rainfall information in the preset range;
inputting the historical relevant information of the upstream hydrological station, the historical relevant information of the branch hydrological station and the historical rainfall information in the preset range into a first neural network model, and complementing the flow information and the rainfall information of the target hydrological station for preset days in the future;
wherein the first neural network model comprises one of: a long and short term memory neural network model, a cyclic neural network model, a multilayer feedforward network model, a convolution network model and a wavelet packet network model.
Optionally, after obtaining the information related to the upstream hydrological station and the information related to the tributary hydrological station, the method further includes:
judging whether the related information of the upstream hydrological station and the related information of the branch hydrological station belong to a preset reasonable range or not;
and when the relevant information of the upstream hydrological station and the relevant information of the tributary hydrological station do not belong to the preset reasonable range, deleting the corresponding relevant information of the upstream hydrological station and the relevant information of the tributary hydrological station.
Optionally, after deleting the relevant information of the corresponding upstream hydrological station and the relevant information of the tributary hydrological station, the method further includes:
and filling the relevant information of the corresponding upstream hydrological station and the relevant information of the tributary hydrological station by adopting a proximity filling method, a linear filling method or a convolution network filling method.
Optionally, predicting future flow information of the target hydrologic station according to the information about the upstream hydrologic station, the information about the tributary hydrologic station, the rainfall information and the time lag information comprises:
inputting the information related to the upstream hydrological station, the information related to the branch hydrological station, the rainfall information and the time lag information into a second neural network model;
taking an output of the second neural network model as future flow information of the target hydrological station;
the second neural network model includes one of: a long and short term memory neural network model, a cyclic neural network model, a multilayer feedforward network model, a convolution network model and a wavelet packet network model.
Optionally, before inputting the information related to the upstream hydrologic station, the information related to the tributary hydrologic station, the rainfall information and the time lag information into a second neural network model, the method further comprises:
carrying out normalization processing on the related information of the upstream hydrological station, the related information of the branch hydrological station, the rainfall information and the time lag information to obtain a normalization processing result;
converting the normalization processing result into a data format of supervised learning by using a pandas shift () function;
and converting the multivariate time series normalization processing result data frame into a data frame suitable for supervised learning by utilizing a series _ to _ superposed () function.
In a second aspect of the present invention, there is provided a river discharge prediction apparatus, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; the designated historical related information of the upstream hydrological station comprises designated historical flow information of the upstream hydrological station and/or designated historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station;
the second acquisition module is used for acquiring the designated historical rainfall information within a preset range from the target hydrological station;
the third acquisition module is used for acquiring the specified historical flow information of the target hydrological station;
the calculation module is used for calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station;
the fourth acquisition module is used for acquiring the related information of the upstream hydrological station and the related information of the branch hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the tributary hydrological station comprises flow information of the tributary hydrological station and/or water level information of the tributary hydrological station;
the fifth acquisition module is used for acquiring rainfall information in the preset range;
and the prediction module is used for predicting the future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
In a third aspect of the present invention, there is provided an electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the river discharge prediction method of any one of the first aspects.
In a fourth aspect of the present invention, there is provided a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the river discharge prediction method according to any one of the first aspect.
The technical scheme of the embodiment of the invention has the following advantages:
the embodiment of the invention provides a river flow prediction method, a device, equipment and a storage medium, wherein the method comprises the following steps: acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; the appointed historical relevant information of the upstream hydrological station comprises appointed historical flow information of the upstream hydrological station and/or appointed historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station; acquiring appointed historical rainfall information within a preset range from a target hydrological station; acquiring appointed historical flow information of a target hydrological station; calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station; acquiring related information of an upstream hydrological station and related information of a tributary hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the branch hydrological station comprises flow information of the branch hydrological station and/or water level information of the branch hydrological station; acquiring rainfall information in a preset range; and predicting the future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information. The problem of low accuracy of the river flow prediction method in the prior art is solved, and the accuracy of river flow prediction is improved.
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 some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a river discharge prediction method according to an embodiment of the invention;
fig. 2 is a block diagram showing the construction of a river discharge predicting apparatus according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.
In the description of the present application, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced device or element must have a particular orientation, be constructed in a particular orientation, and be operated, and thus should not be considered as limiting the present application. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
In this application, the word "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for the purpose of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes are not set forth in detail in order to avoid obscuring the description of the present application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
In accordance with an embodiment of the present invention, there is provided a river discharge prediction method embodiment, it being noted that the steps illustrated in the flow chart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flow chart, in some cases the steps illustrated or described may be performed in an order different than that presented herein.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
In the present embodiment, a river discharge prediction method is provided, and fig. 1 is a flowchart of a river discharge prediction method according to an embodiment of the present invention, as shown in fig. 1, the flowchart includes the following steps:
step S101, acquiring the relevant information of the appointed history of the upstream hydrological station of the target hydrological station and the relevant information of the appointed history of the branch hydrological station. The designated historical related information of the upstream hydrological station may include designated historical flow information of the upstream hydrological station and may further include designated historical water level information of the upstream hydrological station. The specified historical related information of the tributary hydrological station can comprise specified historical flow information of the tributary hydrological station and can also comprise specified historical water level information of the tributary hydrological station. The branch hydrological station is located between the upstream hydrological station and the target hydrological station. Specifically, the spatial relationship of the data needs to be created or determined first. Under different data storage environments, all hydrological stations with upstream and downstream relation and main branch flow relation with a target hydrological station can be determined by selecting modes such as table structure index or station space position observation, and all meteorological rainfall stations in the area can be determined by the space association relation.
Step S102, obtaining appointed historical rainfall information within a preset range from the target hydrological station. Specifically, the rainfall in the preset range influences the flow of the target hydrological station, and all meteorological rainfall stations in the area are determined to monitor rainfall information through the spatial incidence relation.
And step S103, acquiring the specified historical flow information of the target hydrological station. Specifically, the above-described specified historical flow rate information may be acquired by the hydrologic history data of the target hydrologic station.
And step S104, calculating time lag information according to the specified history related information of the upstream hydrological station, the specified history related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station.
Step S105, acquiring relevant information of an upstream hydrological station and relevant information of a tributary hydrological station; the related information of the upstream hydrological station can comprise flow information of the upstream hydrological station and can also comprise water level information of the upstream hydrological station, and the related information of the tributary hydrological station can comprise flow information of the tributary hydrological station and can also comprise water level information of the tributary hydrological station.
And step S106, acquiring rainfall information in the preset range.
And step S107, predicting the future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
In the prior art, a neural network method for predicting river flow based on multi-section water level uniformly numbers last year water level and flow monitoring data of a target hydrological station and water level data of all upstream hydrological stations, integrates data which can be in different years and in the same month/day/hour into a group of independent variables, performs BP neural network fitting on the independent variables and the flow data of the target hydrological station in corresponding time to obtain a neural network model, brings the water level data of a flow time period to be solved into the model when in application to obtain corresponding flow data so as to obtain a water level-flow relation curve, finds out a flow predicted value of the target time period from the curve, but the model for predicting multi-section water level screens and matches the upstream and downstream hydrological and flow data according to an existing date label to form an independent variable data set for simulation, the mode ignores the lag in space and time between the upstream and the downstream of the river, so that the correlation between data is not high, and the data utilization rate and the simulation accuracy are reduced. In another prior art, a river flow prediction method based on a three-dimensional convolutional neural network determines the relationship between factors according to correlation coefficients to obtain a characteristic factor construction data set strongly related to target flow, generates a data sample through characteristics, divides a training set and a verification level, constructs a deep three-dimensional convolutional neural network, and finally uses a model with parameter adjustment and training for prediction.
The current neural network technology for river flow prediction is more accurate in simulation in short time nodes (such as the next 2-4 time nodes), and prediction is difficult in long time nodes.
Through the steps of the embodiment of the invention, firstly, a time sequence data delay analysis method is adopted to determine the delay influence time (namely the time lag information) on the flow of the target hydrological station, and then the time sequence data of each factor is re-matched according to the delay time. Meanwhile, the influence on the downstream target hydrological station is processed in a mode that the upstream main flow hydrological station data, the branch flow hydrological station data and the rainfall data are used as characteristic labels. In some cases, a hydrological station is not arranged on the branch, the flow and water level data cannot be directly acquired, and the branch which cannot directly acquire the flow and water level data can be replaced by adding features into the rainfall data in the target area.
The step S104 mentioned above involves calculating the time lag information from the designation history related information of the upstream hydrological station, the designation history related information of the tributary hydrological station, the designation history rainfall information, and the designation history traffic information of the target hydrological station, specifically, the lag value of the time node t corresponding to the designation history traffic information of the target hydrological station is set to 0, and the lag value of the time node t-n is set to 2n(ii) a Wherein n represents an integer between 1 and | logm |; m is the maximum delay allowed, the appointed historical related information of the upstream hydrological station with the highest relevance with the appointed historical flow information of the target hydrological station is selected from the historical related information of a plurality of upstream hydrological stations corresponding to the time node t-n, the lag value corresponding to the appointed historical related information of the upstream hydrological station is obtained, the first time lag information is calculated according to the lag value corresponding to the appointed historical related information of the upstream hydrological station, the appointed historical related information of the branch hydrological station with the highest relevance with the appointed historical flow information of the target hydrological station is selected from the historical related information of a plurality of branch hydrological stations corresponding to the time node t-n, the lag value corresponding to the appointed historical related information of the branch hydrological station is obtained, the second time lag information is calculated according to the lag value corresponding to the appointed historical related information of the branch hydrological station, and the appointed historical rainfall information of the target hydrological station is selected from the historical rainfall information corresponding to the time node t-n Acquiring the lag value corresponding to the designated historical rainfall information according to the designated historical rainfall information with the highest relevance of the historical flow information, calculating third time lag information according to the lag value corresponding to the designated historical rainfall information, obtaining the time lag information according to the first time lag information, the second time lag information and the third time lag information,for example, the time lag information is obtained by calculating an average value of the first time lag information, the second time lag information and the third time lag information, and those skilled in the art should understand that the calculation method of the time lag information is not limited to the embodiment, and other calculation methods are also within the protection scope of the embodiment according to actual needs.
Specifically, the technical principle of geometric progressive exploration proposed by Sakurai et al in 2005 was adopted in the present embodiment to perform the time delay analysis. The application method is that the lag value of the target time t is 0, the lag value of the last time node t-1 is 2, and the lag values are 0, 2 and 2 in the sample respectively in an exponential mode2，23，24，2|logm|The flow data of the target hydrological station is shifted forward on the time axis by the value and the correlation with other factors such as the flow information and the hydrological information of the upstream hydrological station or the tributary hydrological station is calculated. Finally, the lag value with the highest correlation in the m range is found, and the lag value is the time lag of the target flow data compared with other characteristic data. For example, if it is found that the lag value between the upstream hydrological station data and the target hydrological station data is 2, it represents that the upstream flow state affects the downstream after 2 monitoring times. If the monitoring frequency is once for 4 hours, namely the monitoring value at the downstream has 8 hours of hysteresis on the time axis than the monitoring value at the upstream; or the upstream flow rate regular change has strong correlation with the downstream flow rate regular change after 8 hours.
And performing time delay analysis on all characteristic factor data by using the method, and rearranging the data matching relation according to the lag value. Such as: the delay analysis shows that the delay between upstream and downstream hydrological stations of a river is 12 time nodes (two days if 4 hours and 1 group of monitoring data are available), and then the monitoring data of the upstream 7 months and 1 day is used for fitting the flow value of the downstream 7 months and 3 days. The time stamp of the upstream monitoring data is advanced by 2 days with reference to the time axis of the downstream target hydrological station.
With the above embodiment, for example, data of a target hydrological station after 3 days can be predicted from data of an upstream hydrological station, and if data of the target hydrological station after 7 days is desired to be predicted for a longer time in the future, the data of the target hydrological station needs to be complemented, in an optional embodiment, history related information of the upstream hydrological station, history related information of a tributary hydrological station and history rainfall information within the predetermined range are obtained, the history related information of the upstream hydrological station, the history related information of the tributary hydrological station and the history information within the predetermined range are input to the first neural network model, and flow information and rainfall information of the target hydrological station for a predetermined number of days in the future are complemented; wherein, the quantity of the completion data is: the number of time nodes per day-lag value, wherein the first neural network model comprises one of: a long and short term memory neural network model, a cyclic neural network model, a multilayer feedforward network model, a convolution network model and a wavelet packet network model. Specifically, in a multi-site and multi-variable practical application environment, the data may have different lengths on the time axis after being integrated in the above manner. In practical application, the time range of prediction is 3-7 days according to SD138-85 hydrological information forecast Specification issued by the department of Water conservancy. Therefore, on the premise of reducing a certain accuracy, on one hand, short-term prediction simulation of the flow and the hydrological data of the relevant hydrological site by using a neural network model is required to complete the data length to 7 days. On the other hand, rainfall forecast data needs to be accessed to supplement the meteorological rainfall data for 7 days.
The neural network model is difficult to ensure accuracy in long-term prediction (such as prediction of 56 time nodes in 7 days), and the method reduces the inaccurate influence of the network model in the long-term prediction by reducing the prediction time length of characteristic factor data, changing the prediction data into characteristic factors and the like; meanwhile, the current rainfall forecast accuracy rate of 7 days reaches more than 85%, and the rainfall forecast accuracy rate can be pushed and updated according to a certain frequency, so that the reliability of accessing rainfall forecast data is high.
In addition, if there is no 3-7 day application limit, the step of target hydrology station data completion can be skipped to ensure higher accuracy.
In an optional embodiment, after the relevant information of the upstream hydrological station and the relevant information of the branch hydrological station are obtained, whether the relevant information of the upstream hydrological station and the relevant information of the branch hydrological station belong to a preset reasonable range or not is judged, and when the relevant information of the upstream hydrological station and the relevant information of the branch hydrological station do not belong to the preset reasonable range, the relevant information of the corresponding upstream hydrological station and the relevant information of the branch hydrological station are deleted. Specifically, due to the problem of uneven quality of data in the prior art, abnormal values in hydrological data of each site need to be distinguished and deleted. The methods for judging and deleting include: 1, judging and deleting values exceeding the upper limit and the lower limit of a monitoring instrument; and 2, judging and deleting the value which obviously exceeds the reasonable water quality range. In the deletion mode determined in step 2, it is possible for a layman to set obviously unreasonable deletion such as a value of <0m/s and a value of > 999 m/s; people familiar with local environment and experienced in environmental protection can further set regular expressions to eliminate abnormal data according to the understanding of rivers.
In an optional embodiment, after deleting the relevant information of the corresponding upstream hydrologic station and the relevant information of the tributary hydrologic station, the relevant information of the corresponding upstream hydrologic station and the relevant information of the tributary hydrologic station are filled by adopting a proximity filling method, a linear filling method or a convolutional network filling method. Specifically, after the abnormal data is judged and deleted, the null value needs to be filled, and if the amount of the factor data is small (<3000), methods such as neighbor filling and linear filling can be adopted, wherein the neighbor filling comprises: and (3) replacing the current value with the monitoring value at the closest moment, and linearly filling: obtaining the value of the current moment by linear interpolation by adopting the value of the previous moment and the value of the next moment; if the data size is large (>3000), the data can be padded by a convolution network or the like.
In an optional embodiment, the future flow information of the target hydrologic station is predicted according to the relevant information of the upstream hydrologic station, the relevant information of the branch hydrologic station, the rainfall information and the time lag information, and specifically the future flow information of the target hydrologic station is predicted by inputting the relevant information of the upstream hydrologic station, the relevant information of the branch hydrologic station, the rainfall information and the time lag information into a second neural network model, and the output of the second neural network model is taken as the future flow information of the target hydrologic station. The second neural network model may be a long-short term memory neural network model, a recurrent neural network model, a multi-layer feed-forward network model, a convolutional network model, or a wavelet packet network model.
Before inputting the related information of the upstream hydrological station, the related information of the branch hydrological station, the rainfall information and the time lag information into the second neural network model, in an optional embodiment, the related information of the upstream hydrological station, the related information of the branch hydrological station, the rainfall information and the time lag information are normalized to obtain a normalization processing result, the normalization processing result is converted into a data format of supervised learning by using a pandas shift () function, and a multivariate time series normalization processing result data frame is converted into a data frame suitable for the supervised learning by using a series _ to _ superior () function. Specifically, since the elevation references (i.e., data zeros) of the different site monitoring data may be different, it is necessary to convert to a uniform elevation reference. Integrating data into a single time sequence data set by taking the elevation of a target station as a reference, normalizing the data due to large difference among values of rainfall, flow and water level data, converting the data into a data format for supervised learning by using a pandas shift () function, and finally defining a series _ to _ super () function to convert a multivariable time sequence data frame into a data frame suitable for the supervised learning.
Because the application environment of the embodiment is mainly an automatic information platform, the problem of data real-time pushing exists. In order to ensure that the model can continuously optimize itself along with data pushing, in this embodiment, all time series data are divided into a training set and a test set according to a certain proportion, rather than limiting the number of samples. In practical application, the best application effect at present is to divide the training data in a way of a training set (80%) and a testing set (20%).
In this embodiment, a long-term and short-term memory neural network model is selected as a simulation model. Firstly, reconstructing training set data into a format input model of samples, step lengths and characteristics; and then defining and fitting the model according to the output result. Such as: in this embodiment, in an application scenario of a certain market, by tracking the prediction accuracy, the finally determined model optimal parameters are: a time step (density ═ 1) with 8 features is defined in the output layer; according to the flow prediction target of 3-7 days, the monitoring data (18 groups) of all time nodes in 3 days are selected as the sample number (batch _ size ═ 18) of one training, and MSE and adam are used as loss and optimization functions respectively.
The deployment application is mainly divided into two links: 1, filling the data delay of each station according to the data delay of each factor of each station; and 2, calling an LSTM model for prediction. The calling frequency of the model script is consistent with the pushing frequency of the monitoring data; parameters such as the weight of the characteristic factors and the like can be automatically optimized along with data updating iteration; other parameters of the model suggest a manual parameter adjustment calibration every quarter.
In this embodiment, a river discharge prediction device is further provided, and the device is used to implement the foregoing embodiments and preferred embodiments, and the description of the device is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
The present embodiment provides a river discharge prediction device, as shown in fig. 2, including: a first obtaining module 21, configured to obtain specified history related information of an upstream hydrological station of a target hydrological station and specified history related information of a tributary hydrological station; the appointed historical relevant information of the upstream hydrological station comprises appointed historical flow information of the upstream hydrological station and/or appointed historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station; the second obtaining module 22 is configured to obtain designated historical rainfall information within a predetermined range from the target hydrological station; a third obtaining module 23, configured to obtain specified historical flow information of the target hydrological station; a calculating module 24, configured to calculate time lag information according to the specified history related information of the upstream hydrological station, the specified history related information of the branch hydrological station, the specified historical rainfall information, and the specified historical flow information of the target hydrological station; a fourth obtaining module 25, configured to obtain relevant information of the upstream hydrological station and relevant information of the tributary hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the branch hydrological station comprises flow information of the branch hydrological station and/or water level information of the branch hydrological station; a fifth obtaining module 26, configured to obtain rainfall information within the predetermined range; and the predicting module 27 is used for predicting the future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
The river discharge prediction device in this embodiment is in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory executing one or more software or fixed programs, and/or other devices that can provide the above-described functions.
Further functional descriptions of the modules are the same as those of the corresponding embodiments, and are not repeated herein.
An embodiment of the present invention further provides an electronic device, which has the river discharge prediction apparatus shown in fig. 2.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an electronic device according to an alternative embodiment of the present invention, and as shown in fig. 3, the electronic device may include: at least one processor 301, such as a CPU (Central Processing Unit), at least one communication interface 303, memory 304, and at least one communication bus 302. Wherein a communication bus 302 is used to enable the connection communication between these components. The communication interface 303 may include a Display (Display) and a Keyboard (Keyboard), and the optional communication interface 303 may further include a standard wired interface and a standard wireless interface. The Memory 304 may be a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The memory 304 may optionally be at least one storage device located remotely from the processor 301. Wherein the processor 301 may be combined with the apparatus described in fig. 2, the memory 304 stores an application program, and the processor 301 calls the program code stored in the memory 304 for performing any of the above-mentioned method steps.
The communication bus 302 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 3, but this does not mean only one bus or one type of bus.
The memory 304 may include a volatile memory (RAM), such as a random-access memory (RAM); the memory may also include a non-volatile memory (english: non-volatile memory), such as a flash memory (english: flash memory), a hard disk (english: hard disk drive, abbreviated: HDD) or a solid-state drive (english: SSD); the memory 304 may also comprise a combination of the above-described types of memory.
The processor 301 may be a Central Processing Unit (CPU), a Network Processor (NP), or a combination of a CPU and an NP.
The processor 301 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a field-programmable gate array (FPGA), a General Array Logic (GAL), or any combination thereof.
Optionally, the memory 304 is also used to store program instructions. Processor 301 may invoke program instructions to implement a river discharge prediction method as shown in the embodiment of fig. 1 of the present application.
Embodiments of the present invention further provide a non-transitory computer storage medium, where the computer storage medium stores computer-executable instructions, and the computer-executable instructions may execute the river discharge prediction method in any of the above method embodiments. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A river flow prediction method is characterized by comprising the following steps:
acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; the designated historical related information of the upstream hydrological station comprises designated historical flow information of the upstream hydrological station and/or designated historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station;
acquiring appointed historical rainfall information within a preset range from the target hydrological station;
acquiring appointed historical flow information of the target hydrological station;
calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station;
acquiring relevant information of the upstream hydrological station and relevant information of the tributary hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the tributary hydrological station comprises flow information of the tributary hydrological station and/or water level information of the tributary hydrological station;
acquiring rainfall information in the preset range;
and predicting future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
2. The river discharge prediction method according to claim 1, wherein calculating time lag information from the designation history related information of the upstream hydrological station, the designation history related information of the branch hydrological station, the designation history rainfall information, and the designation history discharge information of the target hydrological station includes:
recording the lag value of a time node t corresponding to the appointed historical flow information of the target hydrological station as 0;
let the lag value of time node t-n be 2n(ii) a Wherein n represents an integer between 1 and | logm |; m is the maximum allowable delay amount;
selecting the designated historical related information of the upstream hydrological station with the highest degree of correlation with the designated historical flow information of the target hydrological station from the historical related information of a plurality of upstream hydrological stations corresponding to the time node t-n;
acquiring a lag value corresponding to the appointed historical relevant information of the upstream hydrological station;
calculating first time lag information according to the lag value corresponding to the appointed historical relevant information of the upstream hydrological station;
selecting the appointed historical related information of the tributary hydrological station with the highest relevance degree from the historical related information of a plurality of tributary hydrological stations corresponding to the time node t-n;
acquiring a lag value corresponding to the appointed historical relevant information of the tributary hydrological station;
calculating second time lag information according to the lag value corresponding to the appointed historical relevant information of the tributary hydrological station;
selecting the appointed historical rainfall information with the highest correlation degree with the appointed historical flow information of the target hydrological station from a plurality of historical rainfall information corresponding to the time node t-n;
acquiring a lag value corresponding to the designated historical rainfall information;
calculating third time lag information according to the lag value corresponding to the designated historical rainfall information;
and obtaining the time lag information according to the first time lag information, the second time lag information and the third time lag information.
3. The river discharge prediction method according to claim 1, characterized in that the method further comprises:
acquiring historical related information of an upstream hydrological station, historical related information of a tributary hydrological station and historical rainfall information in the preset range;
inputting the historical relevant information of the upstream hydrological station, the historical relevant information of the branch hydrological station and the historical rainfall information in the preset range into a first neural network model, and complementing the flow information and the rainfall information of the target hydrological station for preset days in the future;
wherein the first neural network model comprises one of: a long and short term memory neural network model, a cyclic neural network model, a multilayer feedforward network model, a convolution network model and a wavelet packet network model.
4. The river discharge prediction method according to claim 1, after acquiring the information relating to the upstream hydrological station and the information relating to the tributary hydrological station, the method further comprising:
judging whether the related information of the upstream hydrological station and the related information of the branch hydrological station belong to a preset reasonable range or not;
and when the relevant information of the upstream hydrological station and the relevant information of the tributary hydrological station do not belong to the preset reasonable range, deleting the corresponding relevant information of the upstream hydrological station and the relevant information of the tributary hydrological station.
5. The river discharge prediction method according to claim 4, wherein after deleting the relevant information of the corresponding upstream hydrological station and the relevant information of the tributary hydrological station, the method further comprises:
and filling the relevant information of the corresponding upstream hydrological station and the relevant information of the tributary hydrological station by adopting a proximity filling method, a linear filling method or a convolution network filling method.
6. The river discharge prediction method according to claim 1, wherein predicting future discharge information of the target hydrological station based on the information about the upstream hydrological station, the information about the branch hydrological station, the rainfall information, and the time lag information comprises:
inputting the information related to the upstream hydrological station, the information related to the branch hydrological station, the rainfall information and the time lag information into a second neural network model;
taking an output of the second neural network model as future flow information of the target hydrological station;
the second neural network model includes one of: a long and short term memory neural network model, a cyclic neural network model, a multilayer feedforward network model, a convolution network model and a wavelet packet network model.
7. The river discharge prediction method according to claim 6, wherein before inputting the information related to the upstream hydrologic station, the information related to the branch hydrologic station, the rainfall information and the time lag information into a second neural network model, further comprising:
carrying out normalization processing on the related information of the upstream hydrological station, the related information of the branch hydrological station, the rainfall information and the time lag information to obtain a normalization processing result;
converting the normalization processing result into a data format of supervised learning by using a pandas shift () function;
and converting the multivariate time series normalization processing result data frame into a data frame suitable for supervised learning by utilizing series _ to _ super () function.
8. A river discharge prediction device, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring the appointed historical relevant information of an upstream hydrological station of a target hydrological station and the appointed historical relevant information of a tributary hydrological station; the designated historical related information of the upstream hydrological station comprises designated historical flow information of the upstream hydrological station and/or designated historical water level information of the upstream hydrological station; the appointed historical relevant information of the tributary hydrological station comprises appointed historical flow information of the tributary hydrological station and/or appointed historical water level information of the tributary hydrological station;
the second acquisition module is used for acquiring the designated historical rainfall information within a preset range from the target hydrological station;
the third acquisition module is used for acquiring the specified historical flow information of the target hydrological station;
the calculation module is used for calculating time lag information according to the specified historical related information of the upstream hydrological station, the specified historical related information of the branch hydrological station, the specified historical rainfall information and the specified historical flow information of the target hydrological station;
the fourth acquisition module is used for acquiring the related information of the upstream hydrological station and the related information of the branch hydrological station; the related information of the upstream hydrological station comprises flow information of the upstream hydrological station and/or water level information of the upstream hydrological station; the related information of the tributary hydrological station comprises flow information of the tributary hydrological station and/or water level information of the tributary hydrological station;
the fifth acquisition module is used for acquiring rainfall information in the preset range;
and the prediction module is used for predicting the future flow information of the target hydrological station according to the relevant information of the upstream hydrological station, the relevant information of the branch hydrological station, the rainfall information and the time lag information.
9. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the one processor to cause the at least one processor to perform the river discharge prediction method of any one of claims 1-7.
10. A computer readable storage medium having computer instructions stored thereon, wherein the instructions when executed by a processor implement the river discharge prediction method of any one of claims 1 to 7.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011374110 | 2020-11-30 | ||
| CN2020113741106 | 2020-11-30 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113139700A true CN113139700A (en) | 2021-07-20 |
| CN113139700B CN113139700B (en) | 2022-03-11 |
Family
ID=76817967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110545815.8A Active CN113139700B (en) | 2020-11-30 | 2021-05-19 | River flow prediction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113139700B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724889A (en) * | 2021-08-30 | 2021-11-30 | 上海商汤智能科技有限公司 | Prediction method, training method, device and equipment of model thereof and storage medium |
| CN113722094A (en) * | 2021-08-09 | 2021-11-30 | 国网福建省电力有限公司 | Application host resource prediction method based on neural network algorithm |
| CN114091362A (en) * | 2022-01-24 | 2022-02-25 | 江西飞尚科技有限公司 | Water level prediction method, system, storage medium and equipment |
| CN114781766A (en) * | 2022-06-22 | 2022-07-22 | 长江水利委员会长江科学院 | Hydrological information prediction method, device, equipment and storage medium for hydrological site |
| CN115271186A (en) * | 2022-07-18 | 2022-11-01 | 福建中锐网络股份有限公司 | Reservoir water level prediction early warning method based on delay factor and PSO RNN Attention model |
| ES2933392A1 (en) * | 2021-08-06 | 2023-02-06 | Neurite Lab S L | PREDICTION AND EXTRAPOLATION METHOD OF FLOW MEASUREMENTS IN AQUIFER BASINS (Machine-translation by Google Translate, not legally binding) |
| CN116702645A (en) * | 2023-06-07 | 2023-09-05 | 珠江水利委员会珠江水利科学研究院 | Method and system for drawing up warning water level of medium-small river and readable storage medium |
| CN117150974A (en) * | 2023-10-31 | 2023-12-01 | 长江三峡集团实业发展(北京)有限公司 | Reservoir area water surface line prediction method, device, equipment and medium based on large flow |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102032935A (en) * | 2010-12-07 | 2011-04-27 | 杭州电子科技大学 | Soft measurement method for sewage pumping station flow of urban drainage converged network |
| CN103729550A (en) * | 2013-12-18 | 2014-04-16 | 河海大学 | Multi-model integrated flood forecasting method based on propagation time clustering analysis |
| CN106777775A (en) * | 2017-01-10 | 2017-05-31 | 清华大学 | A kind of neural net method based on many section water level forecast river discharges |
| US20180042189A1 (en) * | 2016-08-11 | 2018-02-15 | Rachio, Inc. | Optimized flow control for water infrastructure |
| CN108510072A (en) * | 2018-03-13 | 2018-09-07 | 浙江省水文局 | A kind of discharge of river monitoring data method of quality control based on chaotic neural network |
| CN111079998A (en) * | 2019-12-03 | 2020-04-28 | 华东师范大学 | Flow prediction method based on long and short time sequence correlation attention mechanism model |
-
2021
- 2021-05-19 CN CN202110545815.8A patent/CN113139700B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102032935A (en) * | 2010-12-07 | 2011-04-27 | 杭州电子科技大学 | Soft measurement method for sewage pumping station flow of urban drainage converged network |
| CN103729550A (en) * | 2013-12-18 | 2014-04-16 | 河海大学 | Multi-model integrated flood forecasting method based on propagation time clustering analysis |
| US20180042189A1 (en) * | 2016-08-11 | 2018-02-15 | Rachio, Inc. | Optimized flow control for water infrastructure |
| CN106777775A (en) * | 2017-01-10 | 2017-05-31 | 清华大学 | A kind of neural net method based on many section water level forecast river discharges |
| CN108510072A (en) * | 2018-03-13 | 2018-09-07 | 浙江省水文局 | A kind of discharge of river monitoring data method of quality control based on chaotic neural network |
| CN111079998A (en) * | 2019-12-03 | 2020-04-28 | 华东师范大学 | Flow prediction method based on long and short time sequence correlation attention mechanism model |
Non-Patent Citations (1)
| Title |
|---|
| 陈颖钦 等: "基于相空间重构的水资源时间序列混沌预测模型研究", 《中国水运》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2933392A1 (en) * | 2021-08-06 | 2023-02-06 | Neurite Lab S L | PREDICTION AND EXTRAPOLATION METHOD OF FLOW MEASUREMENTS IN AQUIFER BASINS (Machine-translation by Google Translate, not legally binding) |
| CN113722094A (en) * | 2021-08-09 | 2021-11-30 | 国网福建省电力有限公司 | Application host resource prediction method based on neural network algorithm |
| CN113724889A (en) * | 2021-08-30 | 2021-11-30 | 上海商汤智能科技有限公司 | Prediction method, training method, device and equipment of model thereof and storage medium |
| CN114091362A (en) * | 2022-01-24 | 2022-02-25 | 江西飞尚科技有限公司 | Water level prediction method, system, storage medium and equipment |
| CN114781766A (en) * | 2022-06-22 | 2022-07-22 | 长江水利委员会长江科学院 | Hydrological information prediction method, device, equipment and storage medium for hydrological site |
| CN115271186A (en) * | 2022-07-18 | 2022-11-01 | 福建中锐网络股份有限公司 | Reservoir water level prediction early warning method based on delay factor and PSO RNN Attention model |
| CN115271186B (en) * | 2022-07-18 | 2024-03-15 | 福建中锐网络股份有限公司 | Reservoir water level prediction and early warning method based on delay factor and PSO RNN Attention model |
| CN116702645A (en) * | 2023-06-07 | 2023-09-05 | 珠江水利委员会珠江水利科学研究院 | Method and system for drawing up warning water level of medium-small river and readable storage medium |
| CN116702645B (en) * | 2023-06-07 | 2023-11-21 | 珠江水利委员会珠江水利科学研究院 | Method and system for drawing up warning water level of medium-small river and readable storage medium |
| CN117150974A (en) * | 2023-10-31 | 2023-12-01 | 长江三峡集团实业发展(北京)有限公司 | Reservoir area water surface line prediction method, device, equipment and medium based on large flow |
| CN117150974B (en) * | 2023-10-31 | 2024-01-26 | 长江三峡集团实业发展(北京)有限公司 | Reservoir area water surface line prediction method, device, equipment and medium based on large flow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113139700B (en) | 2022-03-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113139700B (en) | River flow prediction method, device, equipment and storage medium | |
| JP4807565B2 (en) | Flow prediction device | |
| Mejia et al. | Evaluating the effects of parameterized cross section shapes and simplified routing with a coupled distributed hydrologic and hydraulic model | |
| CN113256005B (en) | Power station water level process prediction method and equipment based on neural network model | |
| CN114970377B (en) | Method and system for field flood forecasting based on Xinanjiang and deep learning coupling model | |
| CN114611778B (en) | Reservoir water level early warning method and system based on warehousing flow | |
| KR20140140361A (en) | Water Quality Forecasting of the River Applying Ensemble Streamflow Prediction | |
| CN110265996B (en) | Time characteristic scale modeling method suitable for photovoltaic/wind power prediction | |
| CN113221439B (en) | BP neural network-based drainage system real-time calibration and dynamic prediction method | |
| CN109886461A (en) | A kind of Runoff Forecast method and device | |
| CN113361742A (en) | Hydrologic simulation-based regional comprehensive drought identification method | |
| Jianping et al. | Flood coincidence analysis of Poyang Lake and Yangtze River: risk and influencing factors | |
| Bai et al. | Hydrological probabilistic forecasting based on deep learning and Bayesian optimization algorithm | |
| CN116050595A (en) | Attention mechanism and decomposition mechanism coupled runoff amount prediction method | |
| CN114692981A (en) | Medium-and-long-term runoff forecasting method and system based on Seq2Seq model | |
| CN111815043B (en) | Flood flow prediction method and system based on storm characteristics | |
| CN116881860B (en) | Method and device for forecasting flood of medium and small river based on mathematical statistics analysis | |
| Fattahi et al. | Integrating IHACRES with a data-driven model to investigate the possibility of improving monthly flow estimates | |
| CN107944466A (en) | A kind of rainfall bias correction method based on segmentation thought | |
| CN116796649A (en) | SPEI coarse resolution data space downscaling method and device based on machine learning | |
| CN114911788B (en) | Data interpolation method and device and storage medium | |
| CN111622274A (en) | Method and system for predicting settlement of foundation of high-fill foundation of large grained soil in mountainous area | |
| CN115542427A (en) | Rainfall prediction method and system based on improved neural network model | |
| CN114819264A (en) | Photovoltaic power station irradiance ultra-short term prediction method based on space-time dependence and storage medium | |
| CN114758080A (en) | Sea surface salinity gridding inversion method and device |
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 |