CN115293410A - Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment - Google Patents

Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment Download PDF

Info

Publication number
CN115293410A
CN115293410A CN202210861686.8A CN202210861686A CN115293410A CN 115293410 A CN115293410 A CN 115293410A CN 202210861686 A CN202210861686 A CN 202210861686A CN 115293410 A CN115293410 A CN 115293410A
Authority
CN
China
Prior art keywords
air temperature
temperature prediction
prediction result
random forest
forest classifier
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.)
Withdrawn
Application number
CN202210861686.8A
Other languages
Chinese (zh)
Inventor
黄铜
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.)
3Clear Technology Co Ltd
Original Assignee
3Clear Technology Co Ltd
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 3Clear Technology Co Ltd filed Critical 3Clear Technology Co Ltd
Priority to CN202210861686.8A priority Critical patent/CN115293410A/en
Publication of CN115293410A publication Critical patent/CN115293410A/en
Priority to CN202310073816.6A priority patent/CN116167509B/en
Withdrawn legal-status Critical Current

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
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
    • 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)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Environmental & Geological Engineering (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Ecology (AREA)
  • Environmental Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The air temperature prediction method determines an undetermined air temperature prediction result in a time period to be measured according to the longitude and latitude and the topographic characteristic information, determines a target air temperature prediction result according to weather condition characteristic data of an area to be measured in the time period to be measured and the undetermined air temperature prediction result through a pre-trained air temperature prediction model, and determines a target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result. Therefore, the target air temperature prediction result is determined by the air temperature prediction model according to the terrain feature information and the weather condition feature data of the area to be detected, the accuracy of the target predicted air temperature can be effectively improved, and the influence of small terrain factors and weather conditions on the air temperature is fully considered, so that the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be effectively met.

Description

Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment
Technical Field
The present disclosure relates to the field of weather technologies, and in particular, to a method and an apparatus for predicting air temperature, a storage medium, and an electronic device.
Background
With the development of productivity and the advancement of science and technology, the range of human activities is expanding unprecedentedly, the influence on the nature is also increasing, and accurate weather prediction becomes an indispensable technology in order to further improve productivity. The ground temperature is one of the most concerned meteorological elements in daily life of people, has important influence on the fields of agriculture, industry, service industry and the like, and can directly provide better convenience for industrial and agricultural production and mass life service by accurately predicting the ground temperature. However, in mountainous areas, desert areas and other areas where weather stations are rare, the current air temperature prediction method can only construct general terrain features of climate and distinguish the features from macroscopic non-terrain features, and cannot meet actual demands of microclimate resource investigation and fine agriculture generally.
Disclosure of Invention
An object of the present disclosure is to provide a temperature prediction method, apparatus, storage medium, and electronic device.
In order to achieve the above object, a first aspect of the present disclosure provides a temperature prediction method, including:
acquiring longitude and latitude and topographic feature information of an area to be detected;
determining a prediction result of the to-be-determined air temperature in the time period to be determined according to the longitude and latitude and the topographic characteristic information;
acquiring weather condition characteristic data of the area to be detected in the time period to be detected;
inputting the weather condition characteristic data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model, wherein the air temperature prediction model comprises a plurality of weak classifiers, and different weak classifiers are used for performing air temperature prediction according to characteristic data of different dimensions;
and determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
Optionally, the determining, according to the longitude and latitude and the topographic characteristic information, a prediction result of an undetermined air temperature within a time period to be measured includes:
determining a basic air temperature prediction result through a preset regression function according to the longitude and latitude and the altitude;
determining the plane solar radiation amount according to the longitude and latitude;
determining the solar radiation amount of the slope surface in the time period to be measured according to the longitude and latitude, the slope and the slope direction;
and correcting the basic air temperature prediction result according to the plane solar radiation quantity and the slope solar radiation quantity to obtain the undetermined air temperature prediction result in the time period to be measured.
Optionally, the weather condition feature data includes precipitation, wind speed, and humidity, and the weak classifiers include a first random forest classifier corresponding to a precipitation dimension, a second random forest classifier corresponding to a wind speed dimension, a third random forest classifier corresponding to a humidity dimension, and a fourth random forest classifier corresponding to a prediction result dimension;
the first random forest classifier is used for predicting a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data;
the second random forest classifier is used for predicting a second air temperature prediction result corresponding to the time period to be measured according to the wind speed in the weather condition characteristic data;
the third random forest classifier is used for predicting a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data;
and the fourth random forest classifier is used for predicting a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-measured air temperature prediction result.
Optionally, the inputting the weather condition characteristic data and the undetermined air temperature prediction result into an air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model includes:
simultaneously inputting the weather condition characteristic data and the undetermined air temperature prediction result into the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier so as to obtain a first air temperature prediction result output by the first random forest classifier, a second air temperature prediction result output by the second random forest classifier, a third air temperature prediction result output by the third random forest classifier and a fourth air temperature prediction result output by the fourth random forest classifier;
acquiring target weights corresponding to the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier;
and carrying out weighted summation on the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result and the fourth air temperature prediction result according to the target weight so as to obtain the target air temperature prediction result.
Optionally, the air temperature prediction model is trained by:
acquiring multiple groups of sample data of the area to be detected in multiple preset historical time periods, wherein each group of sample data comprises precipitation sample data, wind speed sample data, humidity sample data, a prediction result of the temperature to be detected in the preset historical time periods and label data of the sample data;
and training a preset Adaboost model by using the multiple groups of sample data as model training data to obtain the air temperature prediction model, wherein the preset Adaboost model comprises a first initial random forest classifier, a second initial random forest classifier, a third initial random forest classifier and a fourth initial random forest classifier, the first initial random forest classifier is used for predicting air temperature according to precipitation sample data, the second initial random forest classifier is used for predicting air temperature according to wind speed sample data, the third initial random forest classifier is used for predicting air temperature according to humidity, and the fourth initial random forest classifier is used for predicting air temperature according to undetermined sample data prediction results in the preset historical time period.
Optionally, the determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result includes:
determining a target air temperature residual value of the area to be measured in the time period to be measured;
and acquiring a target sum of the target air temperature prediction result and the target air temperature residual value, and taking the target sum as the target predicted air temperature.
Optionally, before the determining a target air temperature residual value of the region to be measured in the time period to be measured, the method further includes:
determining target historical air temperature prediction results corresponding to historical sample data of a plurality of target historical time periods through the air temperature prediction model;
obtaining the difference value between the target historical air temperature prediction result and the actually measured air temperature to obtain a plurality of historical air temperature residual values;
processing the plurality of historical air temperature residual values through an interpolation method to obtain an air temperature residual distribution map corresponding to the area to be measured, wherein the air temperature residual distribution map is used for representing the air temperature residual values at different moments;
correspondingly, the determining the target air temperature residual value of the region to be measured in the time period to be measured includes:
and acquiring the target air temperature residual value corresponding to the time period to be measured from the air temperature residual distribution map.
A second aspect of the present disclosure provides an air temperature prediction apparatus, the apparatus comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to acquire longitude and latitude and topographic feature information of an area to be detected;
the first determination module is configured to determine a to-be-determined air temperature prediction result in a to-be-determined time period according to the longitude and latitude and the topographic feature information;
the second acquisition module is configured to acquire weather condition characteristic data of the area to be detected in the time period to be detected;
the second determination module is configured to input the weather condition characteristic data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model so as to obtain a target air temperature prediction result output by the air temperature prediction model, wherein the air temperature prediction model comprises a plurality of weak classifiers, and different weak classifiers are used for performing air temperature prediction according to characteristic data of different dimensions;
a third determination module configured to determine a target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
Optionally, the topographic feature information includes an altitude, a gradient, and a slope of the area to be measured, and the first determining module is configured to:
determining a basic air temperature prediction result through a preset regression function according to the longitude and latitude and the altitude;
determining the plane solar radiation amount according to the longitude and latitude;
determining the solar radiation amount of the slope surface in the time period to be measured according to the longitude and latitude, the slope and the slope direction;
and correcting the basic air temperature prediction result according to the plane solar radiation quantity and the slope solar radiation quantity to obtain the undetermined air temperature prediction result in the time period to be measured.
Optionally, the weather condition feature data includes precipitation, wind speed, and humidity, and the weak classifiers include a first random forest classifier corresponding to a precipitation dimension, a second random forest classifier corresponding to a wind speed dimension, a third random forest classifier corresponding to a humidity dimension, and a fourth random forest classifier corresponding to a prediction result dimension;
the first random forest classifier is used for predicting a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data;
the second random forest classifier is used for predicting a second air temperature prediction result corresponding to the time period to be measured according to the wind speed in the weather condition characteristic data;
the third random forest classifier is used for predicting a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data;
and the fourth random forest classifier is used for predicting a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-determined air temperature prediction result.
Optionally, the second determining module is configured to:
simultaneously inputting the weather condition characteristic data and the undetermined air temperature prediction result into the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier to obtain a first air temperature prediction result output by the first random forest classifier, a second air temperature prediction result output by the second random forest classifier, a third air temperature prediction result output by the third random forest classifier and a fourth air temperature prediction result output by the fourth random forest classifier;
acquiring target weights corresponding to the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier;
and carrying out weighted summation on the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result and the fourth air temperature prediction result according to the target weight so as to obtain the target air temperature prediction result.
Optionally, the apparatus may further comprise a model training module configured to:
acquiring multiple groups of sample data of the area to be detected in multiple preset historical time periods, wherein each group of sample data comprises precipitation sample data, wind speed sample data, humidity sample data, an undetermined air temperature prediction result in the preset historical time periods, and label data of the sample data;
and training a preset Adaboost model by using the multiple groups of sample data as model training data to obtain the air temperature prediction model, wherein the preset Adaboost model comprises a first initial random forest classifier, a second initial random forest classifier, a third initial random forest classifier and a fourth initial random forest classifier, the first initial random forest classifier is used for predicting air temperature according to precipitation sample data, the second initial random forest classifier is used for predicting air temperature according to wind speed sample data, the third initial random forest classifier is used for predicting air temperature according to humidity, and the fourth initial random forest classifier is used for predicting air temperature according to undetermined sample data prediction results in the preset historical time period.
Optionally, the third determining module is configured to:
determining a target air temperature residual value of the area to be measured in the time period to be measured;
and acquiring a target sum of the target air temperature prediction result and the target air temperature residual value, and taking the target sum as the target predicted air temperature.
Optionally, the apparatus further comprises:
a fourth determination module configured to determine, by the air temperature prediction model, target historical air temperature prediction results corresponding to historical sample data of a plurality of target historical time periods;
a third obtaining module configured to obtain a difference between the target historical air temperature prediction result and an actually measured air temperature to obtain a plurality of historical air temperature residual values;
a fifth determining module, configured to process the plurality of historical air temperature residual values through an interpolation method to obtain an air temperature residual distribution map corresponding to the region to be measured, where the air temperature residual distribution map is used to represent the air temperature residual values at different times;
accordingly, the third determination module is configured to:
and acquiring the target air temperature residual value corresponding to the time period to be measured from the air temperature residual distribution map.
A third aspect of the present disclosure provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the first aspect above.
A fourth aspect of the present disclosure provides an electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of the first aspect above.
According to the technical scheme, the to-be-determined air temperature prediction result in the time period to be measured is determined according to the longitude and latitude and the topographic characteristic information, the target air temperature prediction result is determined through a pre-trained air temperature prediction model according to the weather condition characteristic data of the area to be measured in the time period to be measured and the to-be-determined air temperature prediction result, and then the target predicted air temperature of the area to be measured in the time period to be measured is determined according to the target air temperature prediction result. Therefore, a target air temperature prediction result is determined according to the terrain feature information and the weather condition feature data of the area to be measured through the air temperature prediction model, and the final target predicted air temperature is determined according to the target air temperature prediction result, so that the accuracy of the target predicted air temperature can be effectively improved, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be effectively met due to the fact that the influences of small terrain factors and weather conditions on the air temperature are fully considered.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a method of air temperature prediction in accordance with an exemplary embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating a method of air temperature prediction according to the embodiment shown in FIG. 1;
FIG. 3 is a flow chart illustrating another air temperature prediction method according to the embodiment shown in FIG. 1;
FIG. 4 is a flow chart diagram illustrating a method for training an air temperature prediction model in accordance with an exemplary embodiment of the present disclosure;
FIG. 5 is a flow chart illustrating yet another air temperature prediction method according to the embodiment shown in FIG. 1;
FIG. 6 is a block diagram of an air temperature prediction device, shown in an exemplary embodiment of the present disclosure;
FIG. 7 is a block diagram of an air temperature prediction device according to the embodiment shown in FIG. 6;
FIG. 8 is a block diagram illustrating an electronic device in accordance with an exemplary embodiment;
FIG. 9 is a block diagram illustrating another electronic device in accordance with an exemplary embodiment.
Detailed Description
The following detailed description of the embodiments of the disclosure refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
It should be noted that all actions of acquiring signals, information or data in the present disclosure are performed under the premise of complying with the corresponding data protection regulation policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.
Before describing the embodiments of the present disclosure in detail, the following description will first be made on the application scenario of the present disclosure, and the present disclosure can be applied to the air temperature prediction process, especially for the prediction of the air temperature in the area with rare people, bad natural conditions, and rare number of meteorological stations, such as mountain area, desert, and forest. Factors that generally affect the air temperature distribution can be divided into two broad categories: the method mainly comprises the following steps of firstly, macroscopic geographic factors and secondly, microscopic geographic factors, wherein the macroscopic geographic factors mainly show that the air temperature regularly changes along with geographical positions, large terrains and large natural geographic environments; the micro geographic factors mainly refer to gradient, slope direction, small terrain form and the like. Methods for estimating the spatial distribution of air temperature in the related art mainly include the following: the first method is an interpolation method, such as inverse distance weight interpolation, kriging interpolation, etc., which can convert the meteorological station data of discrete points into a continuous data curved surface, and the interpolation method has the disadvantages that: the result is significantly influenced by the number and distribution of the stations, the spatial distribution characteristics of the temperature are mostly not considered, and the influence of the altitude, the terrain and the like is mostly not considered, and the difference change of the temperature under different local terrain conditions is difficult to reflect, so that a large error exists, and the actual requirements of microclimate resource investigation and fine agriculture cannot be met. The second is statistical regression, for example: although the statistical regression method based on the data of the discrete meteorological stations considers the influence of longitude, latitude and altitude on the air temperature, the method is generally rough, can only macroscopically reflect the large trend that the temperature gradually decreases along with the increase of the geographical latitude and altitude, and cannot obtain the air temperature of a local slope area. The third is a temperature terrain correction Model based on a DEM (Digital Elevation Model), which can realize finer temperature distribution by adding small terrain factors such as gradient, slope, terrain shading and the like, but because the terrain factors are unchanged all year round, the temperature at the same time in different years at the same place is the same, which obviously does not accord with the actual situation, so the temperature terrain correction Model also has the situation that the prediction is inaccurate and cannot meet the actual demand. Obviously, the air temperature prediction method in the related art can only construct the characteristic of common zonal distinction of climate and macroscopic non-zonal distinction under the condition that meteorological stations in mountainous areas, desert and the like are rare, and cannot reflect the spatial difference of the outlet air temperature.
In order to solve the technical problems, the present disclosure provides a temperature prediction method, a device, a storage medium and an electronic device, wherein the temperature prediction method determines an undetermined temperature prediction result in a time period to be measured according to the longitude and latitude and the topographic feature information, determines a target temperature prediction result according to weather condition feature data of an area to be measured in the time period to be measured and the undetermined temperature prediction result through a pre-trained temperature prediction model, and determines a target predicted temperature of the area to be measured in the time period to be measured according to the target temperature prediction result. Therefore, the target air temperature prediction result is determined by the air temperature prediction model according to the terrain feature information of the area to be measured and the weather condition feature data, the accuracy of the target predicted air temperature can be effectively improved, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be effectively met due to the fact that the influence of small terrain factors and weather conditions on the air temperature is fully considered.
FIG. 1 is a flow chart illustrating a method of air temperature prediction in accordance with an exemplary embodiment of the present disclosure; as shown in problem 1, the air temperature prediction method may include:
step 101, acquiring longitude and latitude and topographic feature information of an area to be detected.
The topographic feature information may include an altitude, a slope, and a slope of the area to be measured.
And 102, determining a prediction result of the temperature to be determined in the time period to be determined according to the longitude and latitude and the topographic characteristic information.
In this step, one possible implementation manner is: determining a basic air temperature prediction result through a preset regression function according to the longitude and latitude and the altitude; determining the plane solar radiation amount according to the longitude and latitude; determining the solar radiation amount of the slope in the time period to be measured according to the longitude and latitude, the slope and the slope direction; and correcting the basic air temperature prediction result according to the plane solar radiation quantity and the slope solar radiation quantity to obtain the undetermined air temperature prediction result in the time period to be measured.
Another possible implementation is: and training a corresponding machine learning model, and taking the longitude and latitude, the terrain feature information and the time period to be measured as the input of the machine learning model to obtain the prediction result of the undetermined air temperature output by the machine learning model. It should be noted that, the machine learning model may be a neural network model, and a training process of the machine learning model may refer to a training process of a neural network model in the prior art, which is not limited in this disclosure.
Step 103, obtaining weather condition characteristic data of the area to be measured in the time period to be measured.
The weather condition characteristic data may include precipitation, wind speed, and humidity, among others.
And 104, inputting the weather condition characteristic data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model.
The air temperature prediction model comprises a plurality of weak classifiers, and different weak classifiers are used for air temperature prediction according to feature data of different dimensions.
It should be noted that the weak classifiers include a first random forest classifier corresponding to a precipitation dimension, a second random forest classifier corresponding to a wind speed dimension, a third random forest classifier corresponding to a humidity dimension, and a fourth random forest classifier corresponding to a prediction result dimension; the first random forest classifier is used for predicting a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data; the second random forest classifier is used for predicting a second air temperature prediction result corresponding to the time period to be measured according to the air speed in the weather condition characteristic data; the third random forest classifier is used for predicting a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data; and the fourth random forest classifier is used for predicting a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-measured air temperature prediction result. The target air temperature prediction result may be obtained by weighting and summing the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result, and the fourth air temperature prediction result.
And 105, determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
In this step, one possible implementation manner is: the target predicted air temperature is set as the target predicted air temperature.
Another possible implementation is: determining a target air temperature residual value of the area to be measured in the time period to be measured; and acquiring a target sum of the target air temperature prediction result and the target air temperature residual value, and taking the target sum as the target predicted air temperature.
According to the technical scheme, the target air temperature prediction result is determined according to the terrain feature information and the weather condition feature data of the area to be measured through the air temperature prediction model, the final target predicted air temperature is determined according to the target air temperature prediction result, the accuracy of the target predicted air temperature can be effectively improved, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be effectively met due to the fact that the influence of small terrain factors and weather conditions on the air temperature is fully considered.
FIG. 2 is a flow chart of a method of air temperature prediction according to the embodiment shown in FIG. 1; as shown in fig. 2, the step 102 of determining the temperature to be predicted within the time period to be measured according to the longitude and latitude and the topographic feature information in fig. 1 may include the following steps:
and 1021, determining a basic air temperature prediction result through a preset regression function according to the longitude and the latitude and the altitude.
For example, the preset regression function may be a macro air temperature model reflecting general regional characteristics, which is constructed by performing multivariate statistical regression on data collected by existing sites, for example, average air temperature data of a specified region over 10 years may be collected and sorted, and a multivariate first-order regression equation may be constructed:
Figure BDA0003755973170000121
in the above multiple first regression equation, a 0 、a 1 、a 2 、a 3 In order to be the regression coefficient, the method,
Figure BDA0003755973170000122
the latitude, λ, H, and T are the basic air temperature prediction results, the monthly average air temperature data and the longitude, latitude, and altitude of the designated area can be substituted into the multiple linear regression equation to calculate the regression coefficient, so as to obtain the predetermined regression function as:
Figure BDA0003755973170000123
in this step, the longitude, latitude and altitude of the area to be measured are substituted into the preset regression function, so as to obtain the basic air temperature prediction result corresponding to the area to be measured.
And step 1022, determining the plane solar radiation amount according to the longitude and latitude.
Wherein the plane solar radiation quantity S 0flat Can be obtained byThe following equations 1 to 4 are calculated:
Figure BDA0003755973170000124
wherein, in formula 1, the
Figure BDA0003755973170000125
The latitude, omega, of the area to be measured S The sunset hour angle of flat ground, delta is the declination angle of the sun, I 0 Is the solar constant (4.9212 MJ/m 2. H), E 0 A correction factor for the earth orbit.
The solar declination δ can be calculated by the following formula 2:
δ =0.006894-0.399512cos θ +0.072075sin θ -0.006799cos 2 θ +0.00089sin 2 θ equation 2
In the formula 2, θ is a day angle, θ =2 π t/365.2422, t = n-1, and n is a product day.
The sunset time angle omega of the flat ground S Can be calculated by the following equation 3:
Figure BDA0003755973170000131
in the case of the equation 3, the,
Figure BDA0003755973170000132
the latitude of the area to be measured is the declination of the sun.
The earth orbit correction factor can be calculated by the following formula 4:
E 0 =1.000109+0.033494cos theta +0.001472sin theta +0.000768cos2 theta +0.000079sin2 theta equation 4
In this formula 4, θ is the solar angle.
And step 1023, determining the solar radiation amount of the slope surface in the time period to be measured according to the longitude and latitude, the slope and the slope direction.
Wherein, the longitude and latitude, the slope and the slope direction can be extracted from DEM data, and the DEM data is ground to the groundA digitized analog description of the shape. The plane solar radiation quantity S 0slope It can be calculated by formula 4 to formula 8:
Figure BDA0003755973170000133
Figure BDA0003755973170000134
Figure BDA0003755973170000135
Figure BDA0003755973170000136
in the above formulas 5 to 8, α is a gradient, β is a slope direction, n is a discrete number of illumination angles in the time period to be measured,
Figure BDA0003755973170000141
the latitude, omega, of the area to be measured S The sunset hour angle of flat ground, delta is the declination angle of the sun, I 0 Is the sun constant, E 0 As correction factor of the earth orbit, omega r,i The sunrise-hour angle, omega, of the slope corresponding to the ith differential period in the time period to be measured s,i The sunset time angle of the slope surface corresponding to the ith differential time period in the time period to be measured, the sunrise time angle omega r ＝-ω S ，g i Is the topographic shielding degree of the region to be measured, d i Is the shielding degree of the i-th differential period in the period to be measured, the d i Can be calculated by referring to related formulas in the prior art.
And step 1024, correcting the basic air temperature prediction result according to the plane solar radiation amount and the slope solar radiation amount to obtain the to-be-determined air temperature prediction result in the time period to be measured.
In this step, the result of the prediction of the undetermined air temperature can be obtained by calculation according to the formula 9:
Figure BDA0003755973170000142
S 0slope the solar radiation quantity of the slope surface, S 0flat A0, a1, a2 and a3 are regression coefficients of a multiple regression model,
Figure BDA0003755973170000143
to observe the latitude of a site, λ is longitude and H is altitude.
Through the technical scheme from the step 1021 to the step 1024, the influence of longitude, latitude, altitude, local slope and slope direction on the air temperature can be effectively combined, the effect of small topographic factors on the air temperature is fully considered, and more reliable data basis can be provided for subsequent more accurate air temperature prediction.
FIG. 3 is a flow chart illustrating another air temperature prediction method according to the embodiment shown in FIG. 1; as shown in fig. 3, the embodiment of inputting the weather condition characteristic data and the pending air temperature prediction result into the air temperature prediction model to obtain the target air temperature prediction result output by the air temperature prediction model in step 104 in fig. 1 may include the following steps:
step 1041, inputting the weather condition feature data and the undetermined air temperature prediction result into the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier at the same time, so as to obtain a first air temperature prediction result output by the first random forest classifier, a second air temperature prediction result output by the second random forest classifier, a third air temperature prediction result output by the third random forest classifier and a fourth air temperature prediction result output by the fourth random forest classifier.
The first random forest classifier can predict a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data; the second random forest classifier can predict a second air temperature prediction result corresponding to the time period to be measured according to the air speed in the weather condition characteristic data; the third random forest classifier can predict a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data; the fourth random forest classifier can predict a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-measured air temperature prediction result.
1042, obtaining target weights corresponding to the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier.
And the target weight is used for representing the proportion of the output results of the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier in the target air temperature prediction result.
Step 1043, performing weighted summation on the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result and the fourth air temperature prediction result according to the target weight to obtain the target air temperature prediction result.
According to the technical scheme, the precipitation, the wind speed, the humidity and the undetermined air temperature prediction result in the weather condition characteristic data are respectively predicted through a plurality of weak classifiers in the air temperature prediction model, so that a target air temperature prediction result combining the weather condition characteristic data and the small region factor is obtained.
FIG. 4 is a flow chart diagram illustrating a method for training an air temperature prediction model in accordance with an exemplary embodiment of the present disclosure; as shown in fig. 4, the air temperature prediction model can be obtained by training through the following steps S1 to S2:
s1, obtaining multiple groups of sample data of the area to be detected in multiple preset historical time periods.
Each group of the sample data comprises precipitation sample data, wind speed sample data, humidity sample data, a prediction result of the temperature to be determined in the preset historical time period, and labeled data of the sample data, wherein the labeled data can be monitoring data of a meteorological station.
It should be noted that precipitation amount sample data, wind speed sample data, and humidity sample data in each set of sample data may be acquired by monitoring at a meteorological site, and the sample data may be acquired by spatially corresponding the prediction result of the temperature to be determined within a preset historical time period to the original site data acquired at the meteorological site. The original site data acquired from the meteorological site is raster data in a GeoTIFF format, and the original site data comprises: time (Time corresponding to a Time period to be measured), tmean (site temperature), R20_20 (precipitation), fmean (wind speed), umean (humidity), LON (longitude), and LAT (latitude), and the specific operations of spatially corresponding the original site data and the prediction result of the temperature to be measured are as follows:
python programming can be adopted to extract longitude and latitude of meteorological site Data, a geographic affine matrix of a GDAL (geographic Data Abstraction Library) is used to convert the longitude and latitude into a row and column number, a3 x 3 grid is taken from the grid Data in a GeoTIFF format by taking the row and column number as a central point, an average value of each parameter (site temperature, precipitation, wind speed or humidity) in the grid is obtained, and the average value is used as a value corresponding to the longitude and latitude. And finally, adding the value into the site data, wherein the new site data (as a group of sample data) comprises the following sample data: time (Time), tmean (site temperature), R20_20 (precipitation amount), fmean (wind speed), umean (humidity), LON (longitude), LAT (latitude), tmodel _ RF (pending air temperature prediction result), wherein the site temperature is used as the label data corresponding to the current sample data.
And S2, training a preset Adaboost model by taking the multiple groups of sample data as model training data to obtain the air temperature prediction model.
The preset Adaboost model comprises a first initial random forest classifier, a second initial random forest classifier, a third initial random forest classifier and a fourth initial random forest classifier, wherein the first initial random forest classifier is used for predicting air temperature according to precipitation sample data, the second initial random forest classifier is used for predicting air temperature according to wind speed sample data, the third initial random forest classifier is used for predicting air temperature according to humidity sample data, and the fourth initial random forest classifier is used for predicting air temperature according to undetermined air temperature prediction results in a preset historical time period.
It should be noted that the model training data may include a training data set and a testing data set, and each set of sample data in the training data set further includes an initial selected weight for characterizing a probability of being selected as the first target training data set.
When the air temperature prediction model is trained, a first target training data set corresponding to a first initial random forest classifier can be determined from a training data set according to the initial selected weight corresponding to each group of sample data; training the first initial random forest classifier according to the first target training data set to obtain a first to-be-determined random forest classifier; and predicting each group of sample data in the first target training data set through the first to-be-determined random forest classifier, and increasing the initial selection weight corresponding to the sample data under the condition that the prediction result is inconsistent with the marked data so as to obtain the current selection weight of each sample data.
Extracting a second target training data set from the training data set according to the currently selected weight; training the first to-be-determined random forest classifier and the second initial random forest classifier through the second target training data set to obtain a first to-be-determined regression tree model, wherein the first to-be-determined regression tree model comprises the updated first to-be-determined random forest classifier and the updated second to-be-determined random forest classifier, and a first classifier weight corresponding to the updated first to-be-determined random forest classifier and a second classifier weight corresponding to the updated second to-be-determined random forest classifier; and predicting each group of sample data in the second target training data set through the first undetermined regression model, and increasing the current selection weight corresponding to the sample data under the condition that a prediction result is inconsistent with the label data so as to obtain the target selection weight of each group of sample data.
Determining a third target training data set from the training data set according to the target selection weight corresponding to each group of sample data; training the first undetermined regression tree model and the third initial random forest classifier through the third target training data set to obtain a second undetermined regression tree model, wherein the second undetermined regression tree model comprises the updated first undetermined regression model and the updated third undetermined random forest classifier, and the weight corresponding to each random forest classifier; and predicting each group of sample data in the first target training data set through the second undetermined regression model, and increasing the current selection weight corresponding to the sample data under the condition that the prediction result is inconsistent with the tag data so as to obtain the designated selection weight of each group of sample data.
Determining a third target training data set from the training data set through the appointed selected weight, training the second undetermined regression tree model and the third initial random forest classifier through the third target training data set to obtain a third undetermined regression tree model, wherein the third undetermined regression tree model comprises the updated second undetermined regression model and third undetermined random forest classifier, and the weight corresponding to each random forest classifier; predicting each group of sample data in the third target training data set through the third undetermined regression model, increasing the designated selection weight corresponding to the sample data to obtain the final selection weight of each group of sample data under the condition that the prediction result is inconsistent with the label data, determining fourth target training data from the training data set according to the final selection weight, and training the third undetermined regression tree model and the fourth initial random forest classifier through the third target training data set to obtain the air temperature prediction model, wherein the air temperature prediction model comprises the target weight corresponding to each random forest classifier.
After the air temperature prediction model is trained by the training data set, the current air temperature prediction model can be evaluated by the test data set, for example, information such as the degree of fitting, variance, and degree of interpretation of the model on the test data set can be calculated, and when the air temperature prediction model is determined not to satisfy the requirements of the degree of fitting, variance, and degree of interpretation, the training can be repeated until the air temperature prediction model satisfying the requirements of the degree of fitting, variance, and degree of interpretation is obtained.
According to the technical scheme, the Adaboost-based air temperature prediction model is trained, so that the target air temperature prediction result can be efficiently and accurately determined by effectively combining the weather condition characteristics and the small region factors, and the difference change of the air temperature in different weather states under different local terrain conditions can be reflected.
FIG. 5 is a flow chart illustrating yet another air temperature prediction method according to the embodiment shown in FIG. 1; as shown in fig. 5, the determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result in step 105 in fig. 1 may include:
step 1051, determining the target air temperature residual value of the area to be measured in the time period to be measured.
In this step, the air temperature prediction model can be used to determine the target historical air temperature prediction results corresponding to the historical sample data of a plurality of target historical time periods; obtaining the difference value between the target historical air temperature prediction result and the actually measured air temperature to obtain a plurality of historical air temperature residual values; processing the plurality of historical air temperature residual values through an interpolation method to obtain an air temperature residual distribution map corresponding to the area to be measured, wherein the air temperature residual distribution map is used for representing the air temperature residual values at different moments; and acquiring the target air temperature residual value corresponding to the time period to be measured from the air temperature residual distribution map.
Step 1052, obtaining a target sum of the target air temperature prediction result and the target air temperature residual value, and using the target sum as the target predicted air temperature.
According to the technical scheme, the target air temperature prediction result output by the air temperature prediction model can be corrected by combining the actually measured station data, so that more accurate target predicted air temperature can be obtained, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be better met.
FIG. 6 is a block diagram of an air temperature prediction device, shown in an exemplary embodiment of the present disclosure; as shown in fig. 6, the apparatus may include:
a first obtaining module 601 configured to obtain longitude and latitude and topographic feature information of an area to be measured;
a first determination module 602 configured to determine a prediction result of the to-be-determined air temperature in the time period to be determined according to the longitude and latitude and the topographic feature information;
a second obtaining module 603 configured to obtain weather condition characteristic data of the area to be measured in the time period to be measured;
a second determining module 604, configured to input the weather condition feature data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model, where the air temperature prediction model includes a plurality of weak classifiers, and different ones of the weak classifiers are used for performing air temperature prediction according to feature data of different dimensions;
a third determining module 605 configured to determine a target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
According to the technical scheme, the target air temperature prediction result is determined according to the terrain feature information and the weather condition feature data of the area to be measured through the air temperature prediction model, the final target predicted air temperature is determined according to the target air temperature prediction result, the accuracy of the target predicted air temperature can be effectively improved, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be effectively met due to the fact that the influence of small terrain factors and weather conditions on the air temperature is fully considered.
Optionally, the topographic feature information includes an altitude, a gradient, and a slope of the area to be measured, and the first determining module 602 is configured to:
determining a basic air temperature prediction result through a preset regression function according to the longitude and latitude and the altitude;
determining the plane solar radiation amount according to the longitude and latitude;
determining the slope solar radiation amount in the time period to be measured according to the longitude and latitude, the slope and the slope direction;
and correcting the basic air temperature prediction result according to the plane solar radiation quantity and the slope solar radiation quantity to obtain the undetermined air temperature prediction result in the time period to be measured.
Optionally, the weather condition feature data includes precipitation, wind speed and humidity, and the weak classifiers include a first random forest classifier corresponding to a precipitation dimension, a second random forest classifier corresponding to a wind speed dimension, a third random forest classifier corresponding to a humidity dimension and a fourth random forest classifier corresponding to a prediction result dimension;
the first random forest classifier is used for predicting a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data;
the second random forest classifier is used for predicting a second air temperature prediction result corresponding to the time period to be measured according to the air speed in the weather condition characteristic data;
the third random forest classifier is used for predicting a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data;
and the fourth random forest classifier is used for predicting a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-measured air temperature prediction result.
Optionally, the second determining module 604 is configured to:
simultaneously inputting the weather condition characteristic data and the undetermined air temperature prediction result into the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier to obtain a first air temperature prediction result output by the first random forest classifier, a second air temperature prediction result output by the second random forest classifier, a third air temperature prediction result output by the third random forest classifier and a fourth air temperature prediction result output by the fourth random forest classifier;
acquiring target weights corresponding to the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier;
and performing weighted summation on the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result and the fourth air temperature prediction result according to the target weight to obtain the target air temperature prediction result.
FIG. 7 is a block diagram of an air temperature prediction device according to the embodiment shown in FIG. 6; as shown in fig. 7, the apparatus may further include a model training module 606, the model training module 606 configured to:
acquiring multiple groups of sample data of the area to be detected in multiple preset historical time periods, wherein each group of sample data comprises precipitation sample data, wind speed sample data, humidity sample data, an undetermined air temperature prediction result in the preset historical time periods, and label data of the sample data;
and training a preset Adaboost model by taking the multiple groups of sample data as model training data to obtain the air temperature prediction model, wherein the preset Adaboost model comprises a first initial random forest classifier, a second initial random forest classifier, a third initial random forest classifier and a fourth initial random forest classifier, the first initial random forest classifier is used for predicting the air temperature according to precipitation sample data, the second initial random forest classifier is used for predicting the air temperature according to wind speed sample data, the third initial random forest classifier is used for predicting the air temperature according to humidity sample data, and the fourth initial random forest classifier is used for predicting the air temperature according to undetermined air temperature prediction results in the preset historical time period.
According to the technical scheme, the target air temperature prediction result can be efficiently and accurately determined by training the Adaboost-based air temperature prediction model by effectively combining the weather condition characteristics and the small region factors, and the difference change of the air temperature under different weather conditions in different local terrain conditions can be reflected.
Optionally, the third determining module 605 is configured to:
determining a target air temperature residual value of the area to be measured in the time period to be measured;
and acquiring a target sum of the target air temperature prediction result and the target air temperature residual value, and taking the target sum as the target predicted air temperature.
Optionally, the apparatus further comprises:
a fourth determining module 607 configured to determine target historical temperature predictions corresponding to historical sample data of a plurality of target historical time periods through the temperature prediction model;
a third obtaining module 608 configured to obtain a difference between the target historical air temperature prediction result and the measured air temperature to obtain a plurality of historical air temperature residual values;
a fifth determining module 609, configured to process the plurality of historical air temperature residual values through an interpolation method to obtain an air temperature residual distribution map corresponding to the region to be measured, where the air temperature residual distribution map is used to represent the air temperature residual values at different times;
accordingly, the third determination module 605 is configured to:
and acquiring the target air temperature residual value corresponding to the time period to be measured from the air temperature residual distribution map.
According to the technical scheme, the target air temperature prediction result output by the air temperature prediction model can be corrected by combining the actually measured station data, so that more accurate target predicted air temperature can be obtained, and the requirements of microclimate resource investigation and fine agriculture on air temperature prediction can be better met.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
FIG. 8 is a block diagram illustrating an electronic device in accordance with an example embodiment. As shown in fig. 8, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input/output (I/O) interface 704, and a communication component 705.
The processor 701 is configured to control the overall operation of the electronic device 700, so as to complete all or part of the steps of the air temperature prediction method. The memory 702 is used to store various types of data to support operation at the electronic device 700, such as instructions for any application or method operating on the electronic device 700 and application-related data, such as contact data, transmitted and received messages, pictures, audio, video, and so forth. The Memory 702 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically Erasable Programmable Read-Only Memory (EEPROM), erasable Programmable Read-Only Memory (EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia components 703 may include screen and audio components. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in the memory 702 or transmitted through the communication component 705. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 704 provides an interface between the processor 701 and other interface modules, such as a keyboard, mouse, buttons, and the like. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless Communication, such as Wi-Fi, bluetooth, near Field Communication (NFC for short), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, which is not limited herein. The corresponding communication component 705 may thus include: wi-Fi modules, bluetooth modules, NFC modules, and the like.
In an exemplary embodiment, the electronic Device 700 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-described air temperature prediction method.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the air temperature prediction method described above is also provided. For example, the computer readable storage medium may be the above-described memory 702 comprising program instructions executable by the processor 701 of the electronic device 700 to perform the above-described air temperature prediction method.
FIG. 9 is a block diagram illustrating another electronic device in accordance with an example embodiment. For example, electronic device 1900 may be provided as a server. Referring to fig. 9, an electronic device 1900 includes a processor 1922, which may be one or more in number, and a memory 1932 for storing computer programs executable by the processor 1922. The computer program stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processor 1922 may be configured to execute the computer program to perform the air temperature prediction method described above.
Additionally, the electronic device 1900 may also include a power component 1926 and a communication component 1950, the power component 1926 may be configured to perform power management for the electronic device 1900, and the communication component 1950 may be configured to enable communication for the electronic device 1900, e.g., wired or wireless communication. In addition, the electronic device 1900 may also include input/output (I/O) interfaces 1958. The electronic device 1900 may operate based on an operating system, such as Windows Server, stored in memory 1932 TM ，Mac OS X TM ，Unix TM ，Linux TM And so on.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the air temperature prediction method described above is also provided. For example, the non-transitory computer readable storage medium may be the memory 1932 described above that includes program instructions executable by the processor 1922 of the electronic device 1900 to perform the air temperature prediction method described above.
In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-mentioned temperature prediction method when executed by the programmable apparatus.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.

Claims (10)

1. A method of air temperature prediction, the method comprising:
acquiring longitude and latitude and topographic feature information of an area to be detected;
determining a prediction result of the to-be-determined air temperature in the time period to be determined according to the longitude and latitude and the topographic feature information;
acquiring weather condition characteristic data of the area to be detected in the time period to be detected;
inputting the weather condition characteristic data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model, wherein the air temperature prediction model comprises a plurality of weak classifiers, and different weak classifiers are used for performing air temperature prediction according to characteristic data of different dimensions;
and determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
2. The method of claim 1, wherein the topographic feature information includes an altitude, a grade, and a slope of the area to be measured, and wherein determining the temperature prediction to be determined over the time period to be measured based on the longitude and latitude and the topographic feature information comprises:
determining a basic air temperature prediction result through a preset regression function according to the longitude and the latitude and the altitude;
determining the plane solar radiation amount according to the longitude and latitude;
determining the slope solar radiation amount in the time period to be measured according to the longitude and latitude, the slope and the slope direction;
and correcting the basic air temperature prediction result according to the plane solar radiation amount and the slope surface solar radiation amount to obtain the to-be-determined air temperature prediction result in the time period to be measured.
3. The method of claim 1, wherein the weather condition feature data comprises precipitation, wind speed, and humidity, and the plurality of weak classifiers comprises a first random forest classifier corresponding to a precipitation dimension, a second random forest classifier corresponding to a wind speed dimension, a third random forest classifier corresponding to a humidity dimension, and a fourth random forest classifier corresponding to a prediction result dimension;
the first random forest classifier is used for predicting a first air temperature prediction result corresponding to the time period to be measured according to the precipitation in the weather condition characteristic data;
the second random forest classifier is used for predicting a second air temperature prediction result corresponding to the time period to be measured according to the wind speed in the weather condition characteristic data;
the third random forest classifier is used for predicting a third air temperature prediction result corresponding to the time period to be measured according to the humidity in the weather condition characteristic data;
and the fourth random forest classifier is used for predicting a fourth air temperature prediction result corresponding to the time period to be measured according to the to-be-determined air temperature prediction result.
4. The method of claim 3, wherein said inputting said weather condition characteristic data and said pending air temperature prediction into an air temperature prediction model to obtain a target air temperature prediction output by said air temperature prediction model comprises:
simultaneously inputting the weather condition characteristic data and the undetermined air temperature prediction result into the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier so as to obtain a first air temperature prediction result output by the first random forest classifier, a second air temperature prediction result output by the second random forest classifier, a third air temperature prediction result output by the third random forest classifier and a fourth air temperature prediction result output by the fourth random forest classifier;
acquiring target weights corresponding to the first random forest classifier, the second random forest classifier, the third random forest classifier and the fourth random forest classifier;
and carrying out weighted summation on the first air temperature prediction result, the second air temperature prediction result, the third air temperature prediction result and the fourth air temperature prediction result according to the target weight so as to obtain the target air temperature prediction result.
5. The method of claim 1, wherein the air temperature prediction model is trained by:
acquiring multiple groups of sample data of the area to be detected in multiple preset historical time periods, wherein each group of sample data comprises precipitation sample data, wind speed sample data, humidity sample data, a prediction result of the temperature to be detected in the preset historical time periods and label data of the sample data;
and training a preset Adaboost model by using the multiple groups of sample data as model training data to obtain the air temperature prediction model, wherein the preset Adaboost model comprises a first initial random forest classifier, a second initial random forest classifier, a third initial random forest classifier and a fourth initial random forest classifier, the first initial random forest classifier is used for predicting air temperature according to precipitation sample data, the second initial random forest classifier is used for predicting air temperature according to wind speed sample data, the third initial random forest classifier is used for predicting air temperature according to humidity, and the fourth initial random forest classifier is used for predicting air temperature according to undetermined sample data prediction results in the preset historical time period.
6. The method according to claim 1, wherein the determining the target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result comprises:
determining a target air temperature residual value of the area to be measured in the time period to be measured;
and acquiring a target sum of the target air temperature prediction result and the target air temperature residual value, and taking the target sum as the target predicted air temperature.
7. The method of claim 1, wherein prior to said determining a target air temperature residual value for said area under test over said time period under test, said method further comprises:
determining target historical air temperature prediction results corresponding to historical sample data of a plurality of target historical time periods through the air temperature prediction model;
obtaining the difference value between the target historical air temperature prediction result and the actually measured air temperature so as to obtain a plurality of historical air temperature residual values;
processing the plurality of historical air temperature residual values through an interpolation method to obtain an air temperature residual distribution map corresponding to the area to be measured, wherein the air temperature residual distribution map is used for representing the air temperature residual values at different moments;
correspondingly, the determining a target air temperature residual value of the area to be measured in the time period to be measured includes:
and acquiring the target air temperature residual value corresponding to the time period to be measured from the air temperature residual distribution map.
8. An air temperature prediction apparatus, characterized in that the apparatus comprises:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to acquire longitude and latitude and topographic feature information of an area to be detected;
the first determination module is configured to determine a to-be-determined air temperature prediction result in a to-be-determined time period according to the longitude and latitude and the topographic feature information;
the second acquisition module is configured to acquire weather condition characteristic data of the area to be measured in the time period to be measured;
the second determination module is configured to input the weather condition characteristic data and the undetermined air temperature prediction result into a pre-trained air temperature prediction model to obtain a target air temperature prediction result output by the air temperature prediction model, wherein the air temperature prediction model comprises a plurality of weak classifiers, and different weak classifiers are used for carrying out air temperature prediction according to characteristic data with different dimensions;
a third determination module configured to determine a target predicted air temperature of the area to be measured in the time period to be measured according to the target air temperature prediction result.
9. A non-transitory computer-readable storage medium, on which a computer program is stored, which program, when executed by a processor, performs the steps of the method of any one of claims 1 to 7.
10. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the method of any one of claims 1-7.
CN202210861686.8A 2022-07-20 2022-07-20 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment Withdrawn CN115293410A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202210861686.8A CN115293410A (en) 2022-07-20 2022-07-20 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment
CN202310073816.6A CN116167509B (en) 2022-07-20 2023-01-16 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210861686.8A CN115293410A (en) 2022-07-20 2022-07-20 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment

Publications (1)

Publication Number Publication Date
CN115293410A true CN115293410A (en) 2022-11-04

Family

ID=83824868

Family Applications (2)

Application Number Title Priority Date Filing Date
CN202210861686.8A Withdrawn CN115293410A (en) 2022-07-20 2022-07-20 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment
CN202310073816.6A Active CN116167509B (en) 2022-07-20 2023-01-16 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment

Family Applications After (1)

Application Number Title Priority Date Filing Date
CN202310073816.6A Active CN116167509B (en) 2022-07-20 2023-01-16 Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment

Country Status (1)

Country Link
CN (2) CN115293410A (en)

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4792376B2 (en) * 2006-11-15 2011-10-12 三菱重工印刷紙工機械株式会社 Sheet state prediction method
CN106485002B (en) * 2016-10-13 2019-04-19 云南省农业科学院甘蔗研究所 In the method for complicated landform climatic province estimation sugarcane potential production
WO2020209704A1 (en) * 2019-04-10 2020-10-15 임효혁 Method for generating high-resolution marine data, and device using same
CN110599789B (en) * 2019-09-17 2021-04-27 北京心中有数科技有限公司 Road weather prediction method and device, electronic equipment and storage medium
CN110738354B (en) * 2019-09-18 2021-02-05 北京建筑大学 Method and device for predicting particulate matter concentration, storage medium and electronic equipment
CN111639803A (en) * 2020-05-29 2020-09-08 福州市规划设计研究院 Prediction method applied to future vegetation index of area under climate change scene
CN111929748B (en) * 2020-09-17 2022-12-09 南京浦蓝大气环境研究院有限公司 Meteorological element forecasting method and system
CN113159362A (en) * 2020-12-10 2021-07-23 西安天和防务技术股份有限公司 Forest fire grade prediction method and device
CN113222019A (en) * 2021-05-13 2021-08-06 中国南方电网有限责任公司超高压输电公司检修试验中心 Meteorological forecast data processing method, device and equipment for power transmission line tower
CN113569488A (en) * 2021-08-04 2021-10-29 中国科学院地理科学与资源研究所 Somatosensory temperature prediction method and system based on random forest regression
CN113743013A (en) * 2021-09-08 2021-12-03 成都卡普数据服务有限责任公司 XGboost-based temperature prediction data correction method
CN114444020A (en) * 2021-11-09 2022-05-06 国网北京市电力公司 Air temperature forecast correction system for power grid tower point
CN114091764A (en) * 2021-11-24 2022-02-25 中国南方电网有限责任公司超高压输电公司检修试验中心 Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium
CN113988477A (en) * 2021-11-26 2022-01-28 西安化奇数据科技有限公司 Photovoltaic power short-term prediction method and device based on machine learning and storage medium
CN114236643B (en) * 2021-12-17 2023-06-20 中国平安财产保险股份有限公司 Weather forecast method, device, equipment and medium based on weather forecast system
CN114444399A (en) * 2022-02-26 2022-05-06 华北水利水电大学 XGboost-based greenhouse drip irrigation tomato transpiration amount calculation method

Also Published As

Publication number Publication date
CN116167509A (en) 2023-05-26
CN116167509B (en) 2023-09-29

Similar Documents

Publication Publication Date Title
CN111033318A (en) Local weather prediction
Winstral et al. Statistical downscaling of gridded wind speed data using local topography
CN112285807B (en) Meteorological information prediction method and device
CN110705115B (en) Weather forecast method and system based on deep belief network
CN109543907B (en) Complex terrain wind resource assessment method and device
EP3794386B1 (en) Determination of location-specific weather information for agronomic decision support
Biard et al. Automated detection of weather fronts using a deep learning neural network
CN109409014B (en) BP neural network model-based annual illuminable time calculation method
CN115951430A (en) Geomagnetic index forecasting method based on LSTM
CN117008217A (en) Short-time strong precipitation prediction method integrating self-attention module and Unet model
CN111401644A (en) Rainfall downscaling space prediction method based on neural network
Pegion et al. Understanding predictability of daily southeast US precipitation using explainable machine learning
CN112285808B (en) Method for reducing scale of APHRODITE precipitation data
Zhao et al. An objective prediction model for typhoon rainstorm using particle swarm optimization: neural network ensemble
Reinhardt et al. Comparison of different wind data interpolation methods for a region with complex terrain in Central Asia
CN116960962A (en) Mid-long term area load prediction method for cross-area data fusion
Khouider et al. A novel method for interpolating daily station rainfall data using a stochastic lattice model
CN114418243B (en) Distributed new energy cloud grid prediction method and system
CN116340863A (en) Air pollutant prediction method and device, electronic equipment and readable storage medium
CN116167509B (en) Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment
CN115238967A (en) Photovoltaic power prediction method and device combining cloud picture and adjacent power station cluster
Irwin Assessment of the regionalization of precipitation in two Canadian climate regions: A fuzzy clustering approach
CN112668615B (en) Satellite cloud picture prediction method based on depth cross-scale extrapolation fusion
Zhu et al. A method of cloud classification based on DNI
Tsukahara et al. Rainfall forecasting by using residual network with cloud image and humidity

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20221104

WW01 Invention patent application withdrawn after publication