CN116720627B - Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station - Google Patents

Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station Download PDF

Info

Publication number
CN116720627B
CN116720627B CN202310935610.XA CN202310935610A CN116720627B CN 116720627 B CN116720627 B CN 116720627B CN 202310935610 A CN202310935610 A CN 202310935610A CN 116720627 B CN116720627 B CN 116720627B
Authority
CN
China
Prior art keywords
wind
component
data
pressure
weft
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310935610.XA
Other languages
Chinese (zh)
Other versions
CN116720627A (en
Inventor
郎建垒
严允序
玄博元
周颖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Technology
Original Assignee
Beijing University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing University of Technology filed Critical Beijing University of Technology
Priority to CN202310935610.XA priority Critical patent/CN116720627B/en
Publication of CN116720627A publication Critical patent/CN116720627A/en
Application granted granted Critical
Publication of CN116720627B publication Critical patent/CN116720627B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • 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
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Pure & Applied Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Biology (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)

Abstract

The invention discloses a method for predicting a cooperative time sequence of wind speed and wind direction of a monitoring station, belonging to the technical field of atmospheric pollution meteorological element prediction; the method comprises the following steps: acquiring meteorological data of a target site, and decomposing wind speed data and wind direction data; decomposing the air pressure gradient force through wind direction data; screening key influence factors by using the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v; respectively constructing a relation model between the u component and the key influence factor; predicting according to the relation model; and synthesizing the prediction results of the u and v components. The invention also provides a monitoring station wind speed and wind direction collaborative time sequence prediction system. According to the invention, the accurate prediction of wind speed and wind direction is realized by respectively constructing the u-component time sequence prediction model and the v-component time sequence prediction model based on key influence factors, and a foundation can be laid for atmospheric pollution meteorological condition analysis and refined prevention and control for monitoring sites.

Description

Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station
Technical Field
The invention relates to the technical field of atmospheric pollution meteorological element prediction, in particular to a method and a system for predicting a cooperative time sequence of wind speed and wind direction of a monitoring station.
Background
Site contaminant concentration variations are also affected to some extent by meteorological factors, in addition to being directly related to pollution source emissions. Monitoring the contaminant concentration around a site, especially in the local scale range, is very sensitive to meteorological element variations. The wind speed and the wind direction are key factors for driving the diffusion of pollutants, and the direction and the range of the diffusion of the pollutants can be judged by predicting the wind speed and the wind direction of a future period of a station, so that the analysis of pollution causes and the prediction and the judgment of pollution processes are realized, and an environment manager is supported to make feasible pollution prevention and control measures in time. Therefore, the method for quickly and accurately predicting the future wind speed and wind direction time sequence change of the monitoring station has important significance for developing the fine management and control of the atmospheric pollution.
The existing wind speed and wind direction prediction methods mainly comprise two types, wherein one type is based on numerical model prediction, and a prediction result is obtained through a multi-layer grid nesting method or a coupling downscaling model. However, when future prediction is carried out on wind speed and wind direction of a monitoring station in a local scale range, the numerical model cannot well consider the condition of the underlying surface with complex local scale, unavoidable systematic prediction errors exist, the prediction result has large uncertainty, the model has long running period and poor timeliness, and the method is not suitable for carrying out prediction on related elements in the monitoring station. The other class is data-driven model prediction. The existing numerical drive model method mainly builds a prediction model based on a deep learning method and has the advantages of high precision, high timeliness and the like. However, most deep learning models are mainly focused on the prediction of factors such as wind power plants and wind power, less research is conducted on prediction technology of monitoring site wind speed and wind direction, and meanwhile, when site wind speed and wind direction prediction research is conducted, wind speed and wind direction are mostly regarded as an integral variable for prediction, the model structure is simple, analysis and excavation of influences of air pressure on wind speed, wind direction and components thereof in an actual atmosphere movement process are lacked, and therefore unique change characteristics in a site wind factor movement process in an actual local range cannot be captured well, and prediction accuracy is limited. In summary, the existing methods are not suitable for developing time sequence prediction of wind speed and wind direction for monitoring stations, and a new thought and a new method need to be proposed.
Disclosure of Invention
The invention aims to provide a practical monitoring station wind speed and wind direction time sequence prediction method, which realizes accurate prediction of the wind speed and wind direction of a monitoring station in a local scale range and lays a foundation for analysis and refined prevention and control of atmospheric pollution meteorological conditions.
In order to solve the technical problems, the invention provides a method for predicting the cooperative time sequence of wind speed and wind direction of a monitoring station, which comprises the following steps:
acquiring meteorological data of a target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v;
decomposing the air pressure gradient force through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv;
analyzing the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v to obtain key influence factors;
respectively constructing a relation model according to the warp wind component u, the weft wind component v and the key influence factors;
predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
Preferably, the calculation formula of the warp wind component u and the weft wind component v is as follows:
wherein:the wind speed monitoring data of the target site at the moment t; />The wind direction monitoring data of the target station at the moment t; />The warp wind component u of the target station at the corresponding time t; />And the weft wind component v of the target station at the corresponding time t.
Preferably, the pressure sine component Pu and the pressure cosine component Pv are calculated as follows:
wherein: p (P) (t) Pressure monitoring data of a target site at a corresponding time t are represented;representing the pressure sinusoidal component of the target site at the corresponding t moment; />Representing the cosine component of the pressure of the target site at the corresponding time t; />Wind direction monitoring data representing the target site at time t.
Preferably, the meteorological data further comprises temperature data and humidity data.
Preferably, the correlation degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v is analyzed to obtain key influence factors, and the method specifically comprises the following steps:
calculating the association degree between the pressure sine component Pu, the pressure cosine component Pv, the temperature data, the humidity data and the air pressure data and the warp direction wind component u and the weft direction wind component v respectively in a gradual mode;
and (5) carrying out strong correlation or extremely strong correlation screening according to the correlation degree to obtain the key influence factors.
Preferably, the relation model is built according to the warp wind component u, the weft wind component v and the key influence factors, and specifically comprises the following steps:
based on a combined deep learning time sequence prediction algorithm, the relation between the warp wind component u and the weft wind component v and the key influence factors is respectively constructed and used as a relation model.
Preferably, the relationship model is:
wherein:the method comprises the steps of respectively obtaining a warp wind component u, a weft wind component v, a pressure sine component, a pressure cosine component, wind speed data, wind direction data, humidity data, air pressure data and temperature data of a target station at a moment t; />Is a relational model.
Preferably, the formula for synthesizing the warp wind component u prediction result and the weft wind component v prediction result is as follows:
wherein:predicting a result of the warp wind component u of the target station at the moment t; />The method comprises the steps that a weft wind component v prediction result of a target station at a time t is obtained; />The wind speed predicted value of the target site at the time t is obtained; />And the wind direction predicted value of the target station at the corresponding time t is obtained.
Preferably, the pressure gradient force is decomposed through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv, and the method specifically comprises the following steps of:
and decomposing the air pressure gradient force through wind direction data based on two-dimensional vector decomposition, and obtaining a pressure sine component Pu and a pressure cosine component Pv through a proportional transfer method.
The invention also provides a monitoring station wind speed and wind direction cooperative time sequence prediction system, which comprises:
the acquisition module is used for acquiring meteorological data of the target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
the first decomposition module is used for decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v;
the second decomposition module is used for decomposing the air pressure gradient force through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv;
the key influence factor selection module is used for analyzing the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v to obtain a key influence factor;
the relation model construction module is used for respectively constructing a relation model according to the warp wind component u, the weft wind component v and the key influence factors;
the prediction module is used for predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and the synthesis module is used for synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a method for collaborative time sequence prediction of wind speed and wind direction of a monitoring station, which is characterized in that wind vector decomposition is carried out based on a two-dimensional vector decomposition thought, pressure sine and cosine components are obtained by utilizing a proportional transfer method, key influence factors with stronger correlation of u and v components are screened by an influence factor analysis method, a relation model is established by combining a deep learning algorithm, time sequence prediction of u and v components of a target station is respectively carried out based on the model, the u and v components and the key influence factors, and the wind speed and the wind direction at corresponding moments are further synthesized by u and v. According to the invention, the accurate prediction of wind speed and wind direction is realized by respectively constructing the u-component time sequence prediction model and the v-component time sequence prediction model based on key influence factors, and a foundation can be laid for atmospheric pollution meteorological condition analysis and refined prevention and control for monitoring sites.
Drawings
The following describes the embodiments of the present invention in further detail with reference to the accompanying drawings.
FIG. 1 is a flow chart of a method for collaborative timing prediction of wind speed and wind direction at a monitoring station in embodiment 1;
FIG. 2 is a spatial distribution diagram of a monitoring station;
FIG. 3 is a schematic diagram of two-dimensional vector decomposition;
FIG. 4 is an influence factor correlation thermodynamic diagram;
FIG. 5 is a schematic diagram of comparing the wind speed and direction of the target site with the predicted value.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be embodied in many other forms than those herein described, and those skilled in the art will readily appreciate that the present invention may be similarly embodied without departing from the spirit or essential characteristics thereof, and therefore the present invention is not limited to the specific embodiments disclosed below.
The terminology used in the one or more embodiments of the specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the specification. As used in this specification, one or more embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that, although the terms first, second, etc. may be used in one or more embodiments of this specification to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first may also be referred to as a second, and similarly, a second may also be referred to as a first, without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
The invention is described in further detail below with reference to the attached drawing figures:
the invention aims to provide a practical monitoring station wind speed and wind direction collaborative time sequence prediction method, which realizes accurate prediction of the wind speed and the wind direction of a monitoring station in a local scale range and lays a foundation for atmospheric pollution meteorological condition analysis and refined prevention and control.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the invention provides a method for predicting a cooperative time sequence of wind speed and wind direction of a monitoring station, which comprises the following steps:
acquiring meteorological data of a target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v;
decomposing the air pressure gradient force through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv;
analyzing the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v to obtain key influence factors;
respectively constructing a relation model according to the warp wind component u, the weft wind component v and the key influence factors;
predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
Preferably, the calculation formula of the warp wind component u and the weft wind component v is as follows:
wherein:the wind speed monitoring data of the target site at the moment t; />The wind direction monitoring data of the target station at the moment t; />The warp wind component u of the target station at the corresponding time t; />And the weft wind component v of the target station at the corresponding time t.
Preferably, the pressure sine component Pu and the pressure cosine component Pv are calculated as follows:
wherein: p (P) (t) Pressure monitoring data of a target site at a corresponding time t are represented;representing the pressure sinusoidal component of the target site at the corresponding t moment; />Representing the cosine component of the pressure of the target site at the corresponding time t; />Wind direction monitoring data representing the target site at time t.
Preferably, the meteorological data further comprises temperature data and humidity data.
Preferably, the correlation degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v is analyzed to obtain key influence factors, and the method specifically comprises the following steps:
calculating the association degree between the pressure sine component Pu, the pressure cosine component Pv, the temperature data, the humidity data and the air pressure data and the warp direction wind component u and the weft direction wind component v respectively in a gradual mode;
and (5) carrying out strong correlation or extremely strong correlation screening according to the correlation degree to obtain the key influence factors.
Preferably, the relation model is built according to the warp wind component u, the weft wind component v and the key influence factors, and specifically comprises the following steps:
based on a combined deep learning time sequence prediction algorithm, the relation between the warp wind component u and the weft wind component v and the key influence factors is respectively constructed and used as a relation model.
Preferably, the relationship model is:
wherein:respectively at the target sitesA warp direction wind component u, a weft direction wind component v, a pressure sine component, a pressure cosine component, wind speed data, wind direction data, humidity data, air pressure data and temperature data at the moment t; />Is a relational model.
Preferably, the formula for synthesizing the warp wind component u prediction result and the weft wind component v prediction result is as follows:
wherein:predicting a result of the warp wind component u of the target station at the moment t; />The method comprises the steps that a weft wind component v prediction result of a target station at a time t is obtained; />The wind speed predicted value of the target site at the time t is obtained; />And the wind direction predicted value of the target station at the corresponding time t is obtained.
Preferably, the pressure gradient force is decomposed through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv, and the method specifically comprises the following steps of:
and decomposing the air pressure gradient force through wind direction data based on two-dimensional vector decomposition, and obtaining a pressure sine component Pu and a pressure cosine component Pv through a proportional transfer method.
The invention also provides a monitoring station wind speed and wind direction cooperative time sequence prediction system, which comprises:
the acquisition module is used for acquiring meteorological data of the target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
the first decomposition module is used for decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v;
the second decomposition module is used for decomposing the air pressure gradient force through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv;
the key influence factor selection module is used for analyzing the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v to obtain a key influence factor;
the relation model construction module is used for respectively constructing a relation model according to the warp wind component u, the weft wind component v and the key influence factors;
the prediction module is used for predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and the synthesis module is used for synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
The invention provides a method for collaborative time sequence prediction of wind speed and wind direction of a monitoring station, which is characterized in that wind vector decomposition is carried out based on a two-dimensional vector decomposition thought, pressure sine and cosine components are obtained by utilizing a proportional transfer method, key influence factors with stronger correlation of u and v components are screened by an influence factor analysis method, a relation model is established by combining a deep learning algorithm, time sequence prediction of u and v components of a target station is respectively carried out based on the model, the u and v components and the key influence factors, and the wind speed and the wind direction at corresponding moments are further synthesized by u and v. According to the invention, the accurate prediction of wind speed and wind direction is realized by respectively constructing the u-component time sequence prediction model and the v-component time sequence prediction model based on key influence factors, and a foundation can be laid for atmospheric pollution meteorological condition analysis and refined prevention and control for monitoring sites.
In order to better illustrate the technical effects of the present invention, the present invention provides the following specific embodiments to illustrate the above technical flow:
embodiment 1, a practical method for predicting a cooperative time sequence of wind speed and wind direction of a monitoring station, as shown in fig. 1, includes the following steps:
acquiring wind speed and wind direction monitoring data of a target site, and carrying out wind vector decomposition to obtain a warp direction wind component u and a weft direction wind component v at corresponding moments;
based on a proportional transfer method, decomposing the air pressure gradient force, and further obtaining sine and cosine components Pu and Pv of the target site pressure, wherein the sine and cosine components Pu and Pv are used as potential influence factors for influencing u and v components;
screening key influence factors with strong correlation with u and v components by using an influence factor analysis method;
based on a deep learning time sequence prediction algorithm, respectively constructing a relation model between the u component and the key influence factor;
combining deep learning based on a relation model between u and v components and key influence factors, and respectively carrying out time sequence prediction of a target site u and v components for a certain time effect length in the future;
and synthesizing the u and v components by using a mathematical method to obtain a wind speed and wind direction predicted value at the target moment.
Further, in the step (1), wind vector decomposition specifically means two-dimensional vector decomposition is performed on wind elements by using a Cartesian rectangular coordinate system, so as to obtain a monitoring sequence of a warp wind component u and a weft wind component v at a moment corresponding to wind element monitoring data. The specific calculation formula is as follows:
wherein:the wind speed monitoring data of the target site at the moment t; />The wind direction monitoring data of the target station at the moment t; />The warp wind component u of the target station at the corresponding time t; />And the weft wind component v of the target station at the corresponding time t.
Further, the proportional transfer method in the step (2) specifically refers to that in the local range, the earth gravity and the friction force can be regarded as known quantities or invariants, and the atmospheric vector motion equation and the pressure gradient force calculation formula are combined to obtain the fact that the components u and v of the monitoring stations in the local scale range are in direct proportion to the pressure, so that the sine and cosine components of the pressure are obtained by decomposing the pressure gradient force. The specific relation is as follows:
wherein: v and u respectively represent the wind speed in the weft direction and the warp direction; f is a ground deflection force parameter;is air density; p is the atmospheric pressure.
F (t) = P (t) S
F in the formula (t) Respectively representing the air pressure of the target site at the time tGradient force, fu (t) 、Fv (t) Respectively representing the component of the air pressure gradient force of the target site at the time t,the air pressure monitoring value of the station at the time t is shown, and S represents the area;
wherein: p (P) (t) Pressure monitoring data of a target site at a corresponding time t are represented;representing the pressure sinusoidal component of the target site at the corresponding t moment; />Representing the cosine component of the pressure of the target site at the corresponding time t; />Wind direction monitoring data representing the target site at time t.
Further, the influence factor analysis method in the step (3) specifically refers to screening influence factors which are strongly or extremely strongly related to u and v components and serve as key influence factors for influencing the prediction effect of the u and v components based on the correlation between the u and v components and the u and v monitoring sequences, the Pu and Pv monitoring sequences and the time-by-time monitoring data of meteorological elements (such as temperature, humidity, wind speed, wind direction and air pressure), so that complex relations between wind speed and wind direction in the atmospheric environment and other meteorological elements are reflected, the capture of time sequence change characteristics of the wind elements of a monitoring station by a model is improved, and the prediction accuracy of the wind speed and the wind direction is improved.
Further, the relation model between the u and v components and the key influence factors in the step (4) specifically refers to the key influence factors obtained in the step (3), and the relation between the u and v components and the key influence factors is respectively constructed by combining a deep learning time sequence prediction algorithm. The specific relation is as follows:
wherein:the method comprises the steps of respectively obtaining a warp wind component u, a weft wind component v, a pressure sine component, a pressure cosine component, wind speed data, wind direction data, humidity data, air pressure data and temperature data of a target station at a moment t; />And establishing a relation model for the deep learning model based on the u and v components and key influence factors established by the monitoring station at the time t.
Further, in the step (5), the relation model based on the u and v components and the key influence factors in combination with deep learning specifically means that a target data set (historical multi-year u and v monitoring sequences, pu and Pv monitoring sequences, temperature, humidity, wind speed, wind direction and pressure monitoring data) is set according to a certain proportion to perform training, and after the training is performed, time sequence prediction of a certain future aging length is performed on the u and v components of the target site.
In step (6), a mathematical method is adopted to specifically refer to that a predicted sequence of u and v components is converted by a trigonometric function to obtain a wind speed predicted value at a corresponding moment through a vector synthesis formula. The conversion formula is as follows:
;
wherein:predicting a result of the warp wind component u of the target station at the moment t; />The method comprises the steps that a weft wind component v prediction result of a target station at a time t is obtained; />The wind speed predicted value of the target site at the time t is obtained; />And the wind direction predicted value of the target station at the corresponding time t is obtained.
Example 1: based on the prediction method of the invention, 3 monitoring stations randomly selected in the encryption observation network in the Linyi city of Shandong province are taken as research objects, and short-term (24 h) wind speed and wind direction time sequence prediction of each monitoring station of which the month is represented by 2022 years 1, 4, 7 and 10 months is taken as research objects, and the specific process is as follows:
in the case of the present embodiment, 3 monitoring sites in Yi city in Shandong province are randomly selected as research objects and marked as national control site S1, provincial control site S2 and municipal control site S3 (as shown in FIG. 2).
Time-by-time monitoring data (2021, 1, 10, 31, 2022) of each target station are obtained, a two-dimensional vector decomposition thought is adopted, the time-by-time wind direction monitoring data of each monitoring station 2021, 1, 2022, 10, 31 are taken as wind direction angles, and two-dimensional vector decomposition is carried out according to a Cartesian rectangular coordinate system (as shown in FIG. 3) to respectively obtain corresponding time-by-time warp wind components u and weft wind components v.
And (2) based on an atmospheric vector motion equation, regarding the earth gravitation and the friction force in the local scale range as known quantities or invariants, so as to obtain the influence of the air pressure gradient force on the u and v component changes of the target site in the local scale range, wherein the u and v component sizes are in direct proportion to the air pressure gradient force, the air pressure gradient force generated around the site is obtained in direct proportion to the pressure based on an air pressure gradient force calculation formula, the sine and cosine values of the u and v component and the air pressure are further obtained in direct proportion through a proportion transmission principle, and the time-by-time monitoring sequence of the 1 month 1 day 2022 day 31 to the time-by-time monitoring sequence of the weft air pressure component Pv is calculated and used as one of potential influence factors for influencing the u and v components.
The specific formula is as follows:
wherein: v is a wind speed vector; g is gravity acceleration;is air density; p is atmospheric pressure; />Is a Hamilton operator; />Is the pressure gradient force; />Is the rotation angular velocity of the earth; />Is a ground deflection force; f is molecular friction; t is time.
Wherein: v and u respectively represent the wind speed in the weft direction and the warp direction; f is a ground deflection force parameter;is air density; p is the atmospheric pressure. />、/>Respectively represent the weft and warp friction forces.
Wherein: v and u respectively represent the wind speed in the weft direction and the warp direction; f is a ground deflection force parameter;is air density; p is the atmospheric pressure.
F (t) = P (t) S
F in the formula (t) Fu respectively representing the air pressure gradient force of the target site at the time t (t) 、Fv (t) Respectively representing the component of the air pressure gradient force of the target site at the time t,the air pressure monitoring value of the station at the time t is shown, and S represents the area;
wherein: p (P) (t) Pressure monitoring data representing target site at corresponding t moment；Representing the pressure sinusoidal component of the target site at the corresponding t moment; />Representing the cosine component of the pressure of the target site at the corresponding time t; />Wind direction monitoring data representing the target site at time t.
And (3) taking meteorological factors such as temperature, humidity, wind direction, wind speed, air pressure and Pu, pv, u, v as influence factors, performing correlation thermodynamic diagram analysis by using an influence factor analysis method to screen key influence factors influencing the predicted effect of u and v components of target sites (S1, S2 and S3), selecting characteristic variables which are strongly or extremely strongly correlated with the u and v components as key influence factors of the u and v components respectively, and finally selecting Pu and Pv as key influence factors according to the correlation thermodynamic diagram analysis result (shown in figure 4).
Step (4): combining the monitoring sequences of the u and v components with the Pu and Pv monitoring sequences and a deep learning time sequence prediction algorithm (LSTM), and establishing the relation between the u and v and the characteristic variables:
and (5) selecting the time-by-time monitoring data of each target site (S1, S2 and S3) in the previous year as a data set, taking 80% of the data in the front of the data set as a training set and 20% of the data in the rear of the data set as a test set for training, and respectively carrying out time sequence prediction of the time-by-time u and v components of each site 2022 in the future of 24 hours of 1, 4, 7 and 10 months.
Step (6): based on the predicted sequences of u and v components of each target site 2022, 1, 4, 7 and 10 months in future 24 hours time by time, which are obtained by the u and v component time sequence prediction model, wind speed predicted values at corresponding moments are obtained through vector synthesis formula conversion.
And simultaneously, obtaining a wind direction predicted value at a corresponding moment by adopting a trigonometric function method.
Through the wind speed (left in fig. 5) and wind direction (right in fig. 5) prediction results and actual monitoring trend conditions of the future 24h step sizes in the study period of the S1, S2 and S3 sites, the wind speed and wind direction collaborative time sequence prediction model can better reflect the unique characteristics of the time sequence changes of the wind speed and the wind direction of the sites, and accurate prediction of the wind speed and the wind direction is realized.
The wind speed and wind direction prediction performance evaluation indexes of each site are shown in table 1, and the wind speed and wind direction collaborative time sequence prediction model has small fluctuation of the wind speed and wind direction prediction performance of different sites, and has good robustness.
In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the embodiments described above are merely illustrative, and the division of modules, or units is merely a logical division of functionality, and there may be additional divisions of actual implementations, e.g., multiple units, modules, or components may be combined or integrated into another device, or some features may be omitted, or not performed.
The units may or may not be physically separate, and the components shown as units may be one physical unit or a plurality of physical units, may be located in one place, or may be distributed in a plurality of different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such embodiments, the computer program may be downloaded and installed from a network via a communication portion, and/or installed from a removable medium. The above-described functions defined in the method of the present invention are performed when the computer program is executed by a Central Processing Unit (CPU). The computer readable medium of the present invention may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the present invention is not limited thereto, but any changes or substitutions within the technical scope of the present invention should be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (3)

1. A method for predicting the cooperative time sequence of wind speed and wind direction of a monitoring station is characterized by comprising the following steps:
acquiring meteorological data of a target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v; the calculation formulas of the warp wind component u and the weft wind component v are as follows:
u obs(t) ＝-V t sin(θ t )
v obs(t) ＝-V t cos(θ t )
wherein: v (V) t The wind speed monitoring data of the target site at the moment t; θ t The wind direction monitoring data of the target station at the moment t; u (u) obs(t) The warp wind component u of the target station at the corresponding time t; v obs(t) The weft wind component v of the target station at the corresponding time t;
decomposing the air pressure gradient force through wind direction data based on two-dimensional vector decomposition, and obtaining a pressure sine component Pu and a pressure cosine component Pv through a proportional transfer method; the pressure sine component Pu and the pressure cosine component Pv are calculated as follows:
F (t) ＝P (t) S
Fu (t) ＝-SP (t) sin(θ (t) )
Fv (t) ＝-SP (t) cos(θ (t) )
wherein: f (F) (t) Fu respectively representing the air pressure gradient force of the target site at the time t (t) 、Fv (t) Respectively representing component force of air pressure gradient force of a target site at the time t, wherein S represents area;
Pu (t) ＝-P (t) sin(θ (t) )
Pv (t) ＝-P (t) cos(θ (t) )
wherein: p (P) (t) Pressure monitoring data of a target site at a corresponding time t are represented; pu (Pu) (t) Representing the pressure sinusoidal component of the target site at the corresponding t moment; pv (Pv) (t) Representing the cosine component of the pressure of the target site at the corresponding time t; θ (t) Wind direction monitoring data representing a target site at a time t;
calculating the association degree between the pressure sine component Pu, the pressure cosine component Pv, the temperature data, the humidity data and the air pressure data and the warp direction wind component u and the weft direction wind component v respectively in a gradual mode;
based on the correlation between the warp wind component u, the weft wind component v, the u and v monitoring sequences, the Pu and Pv monitoring sequences and the meteorological element time-by-time monitoring data, screening influence factors which are strongly correlated with the u and v components to be respectively used as key influence factors for influencing the prediction effect of the u and v components; the meteorological elements comprise temperature, humidity, wind speed, wind direction and air pressure;
based on a combined deep learning time sequence prediction algorithm, respectively constructing relations between the warp wind component u, the weft wind component v and key influence factors as relation models; the relation model is as follows:
Yu&Yv (t) ＝(u obs(t) v obs(t) Pu (t) Pv (t) WS (t) WD (t) RH (t) PRS (t) T (t) )′
wherein: u (u) obs(t) 、v obs(t) 、Pu (t) 、Pv (t) 、WS (t) 、WD (t) 、RH (t) 、PRS (t) 、T (t) The warp direction wind component u, the weft direction wind component v and the pressure sine component of the target station at the time t are respectivelyThe system comprises a pressure cosine component, wind speed data, wind direction data, humidity data, air pressure data and temperature data; yu (Yu)&Yv (t) Is a relational model;
predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
2. The method for collaborative timing prediction of wind speed and wind direction at a monitoring station according to claim 1, wherein the formula for combining the predicted result of the warp wind component u and the predicted result of the weft wind component v is as follows:
WD’ (t) ＝arctan2(v,u)
wherein: v pred(t) Predicting a result of the warp wind component u of the target station at the moment t; u (u) pred(t) The method comprises the steps that a weft wind component v prediction result of a target station at a time t is obtained; WS'. (t) The wind speed predicted value of the target site at the time t is obtained; WD' (t) And the wind direction predicted value of the target station at the corresponding time t is obtained.
3. A monitoring site wind speed and wind direction cooperative timing prediction system for implementing the monitoring site wind speed and wind direction cooperative timing prediction method according to any one of claims 1-2, comprising:
the acquisition module is used for acquiring meteorological data of the target site; the meteorological data comprise wind speed data, wind direction data and air pressure data;
the first decomposition module is used for decomposing the wind speed data and the wind direction data to obtain a warp wind component u and a weft wind component v;
the second decomposition module is used for decomposing the air pressure gradient force through wind direction data to obtain a pressure sine component Pu and a pressure cosine component Pv;
the key influence factor selection module is used for analyzing the association degree of the pressure sine component Pu, the pressure cosine component Pv and the meteorological data with the warp direction wind component u and the weft direction wind component v to obtain a key influence factor;
the relation model construction module is used for respectively constructing a relation model according to the warp wind component u, the weft wind component v and the key influence factors;
the prediction module is used for predicting according to the relation model to obtain a predicted result of the warp wind component u and a predicted result of the weft wind component v;
and the synthesis module is used for synthesizing the predicted result of the warp wind component u and the predicted result of the weft wind component v to obtain a wind speed predicted value and a wind direction predicted value.
CN202310935610.XA 2023-07-28 2023-07-28 Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station Active CN116720627B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310935610.XA CN116720627B (en) 2023-07-28 2023-07-28 Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310935610.XA CN116720627B (en) 2023-07-28 2023-07-28 Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station

Publications (2)

Publication Number Publication Date
CN116720627A CN116720627A (en) 2023-09-08
CN116720627B true CN116720627B (en) 2024-03-19

Family

ID=87871759

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310935610.XA Active CN116720627B (en) 2023-07-28 2023-07-28 Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station

Country Status (1)

Country Link
CN (1) CN116720627B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117852897A (en) * 2023-11-17 2024-04-09 中国人民解放军63796部队 Screening method and system for key factors generated by typhoons

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013120395A1 (en) * 2012-02-13 2013-08-22 国家电网公司 Ultra-short-term prediction method comprising upstream/downstream effect real-time monitoring
CN113326969A (en) * 2021-04-29 2021-08-31 淮阴工学院 Short-term wind speed prediction method and system based on improved whale algorithm for optimizing ELM
CN114897204A (en) * 2022-02-17 2022-08-12 上海电机学院 Method and device for predicting short-term wind speed of offshore wind farm
CN115494562A (en) * 2022-09-02 2022-12-20 浙大城市学院 Indirect wind direction prediction method considering wind speed and wind direction correlation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013120395A1 (en) * 2012-02-13 2013-08-22 国家电网公司 Ultra-short-term prediction method comprising upstream/downstream effect real-time monitoring
CN113326969A (en) * 2021-04-29 2021-08-31 淮阴工学院 Short-term wind speed prediction method and system based on improved whale algorithm for optimizing ELM
CN114897204A (en) * 2022-02-17 2022-08-12 上海电机学院 Method and device for predicting short-term wind speed of offshore wind farm
CN115494562A (en) * 2022-09-02 2022-12-20 浙大城市学院 Indirect wind direction prediction method considering wind speed and wind direction correlation

Also Published As

Publication number Publication date
CN116720627A (en) 2023-09-08

Similar Documents

Publication Publication Date Title
CN110085281B (en) Environmental pollution traceability system and method based on characteristic pollution factor source analysis
CN116720627B (en) Collaborative time sequence prediction method and system for wind speed and wind direction of monitoring station
Harilal et al. Augmented convolutional LSTMs for generation of high-resolution climate change projections
Alobaidi et al. Mapping of the solar irradiance in the UAE using advanced artificial neural network ensemble
CN117116382B (en) Water quality space-time prediction method and system for water-bearing lake under influence of diversion engineering
Zhang et al. Machine learning‐based modeling of vegetation leaf area index and gross primary productivity across North America and comparison with a process‐based model
Song et al. A continuous space location model and a particle swarm optimization-based heuristic algorithm for maximizing the allocation of ocean-moored buoys
CN116644689B (en) Method and system for strong and rapid back calculation of atmospheric pollution source of local scale under complex underlying surface
CN108764527B (en) Screening method for soil organic carbon library time-space dynamic prediction optimal environment variables
CN113125635A (en) Atmospheric pollution early warning method and device and readable storage medium
CN115471679A (en) Method and intelligent system for synchronously assimilating water level and flow of natural river
Niu et al. Automatic quality control of crowdsourced rainfall data with multiple noises: A machine learning approach
Nathaniel et al. Chaosbench: A multi-channel, physics-based benchmark for subseasonal-to-seasonal climate prediction
CN117671504A (en) Marine wind power identification method and system based on yolo algorithm
Kido et al. JCOPE-FGO: an eddy-resolving quasi-global ocean reanalysis product
CN114814135B (en) River water quality pollution tracing method and system based on multivariate monitoring
CN115964869A (en) High-space-time resolution atmospheric pollution diffusion migration simulation method
CN116009118A (en) Method and system for predicting short-term atmospheric optical turbulence profile based on artificial intelligence
Datta‐Barua et al. Lower thermospheric material transport via Lagrangian coherent structures
Yongping et al. Microphysical retrieval from Doppler radar reflectivity using variational data assimilation
Huntzinger et al. The utility of continuous atmospheric measurements for identifying biospheric CO2 flux variability
Sokolowsky et al. tobac v1. 5: Introducing fast 3D tracking, splits and mergers, and other enhancements for identifying and analysing meteorological phenomena
CN114527235A (en) Real-time quantitative detection method for emission intensity
Hellan et al. Optimising placement of pollution sensors in windy environments
Michalak Adapting a fixed-lag Kalman smoother to a geostatistical atmospheric inversion framework

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