Calibration method for short-term rainfall forecast and computer-readable storage medium

Abstract

The present invention discloses a calibration method for short-term rainfall forecasts and a computer-readable storage medium. The method comprises: determining a historical time point corresponding to historical radar echo data associated with the latest radar echo data as a first historical time point; determining a first historical time point corresponding to historical radar echo sequence data associated with the latest radar echo sequence data as a second historical time point; determining a second historical time point corresponding to historical measured rainfall sequence data having the highest correlation with the measured rainfall sequence data as a third historical time point; and calibrating short-term rainfall forecast data for the forecast area within a preset second time period after the latest time point based on measured rainfall data for the forecast area within a preset second time period after the third historical time point. The present invention can improve the accuracy of short-term rainfall forecasts.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological forecasting, and in particular to a calibration method for short-term rainfall forecast and a computer-readable storage medium. Background Art

[0002] For short-term rainfall forecasts, we can only predict the possibility of severe convective weather, but cannot predict the specific areas where the rainfall it brings will fall, which has great uncertainty.

[0003] Traditional methods for short-term rainfall forecasting include statistical precipitation forecasting and weather radar extrapolation. Statistical precipitation forecasting is based on statistical principles and employs probability theory and mathematical statistics to predict rainfall. Weather radar extrapolation is primarily used for near-term rainfall forecasting and storm monitoring. It detects the location, shape, intensity, and direction of cloud and precipitation by measuring echo signal strength, frequency, and phase. The intensity of the radar echo reflects the intensity of the cloud or precipitation, while the frequency and phase reflect its speed and direction. Generally speaking, the larger and more numerous the raindrops, the stronger the reflected signal. Based on the echo images observed by the radar at a previous moment and the current moment, the echo's speed and direction are calculated. This speed and direction are then extrapolated to predict the echo's position at a future moment.

[0004] However, statistical forecasting methods require extensive historical data, are formulaic, and require a consistent relationship between the selected factors and the predicted quantity. This limits the effectiveness of forecasts and the degree to which the relationships between these factors and the predicted quantity align with historical meteorological data, limiting the stability of short-term rainfall forecasts. Using meteorological radar extrapolation to analyze radar echo data, however, is affected by geographic factors and the accuracy of radar data. Simple linear extrapolation forecasting methods can result in large forecast errors and low accuracy in situations with significant weather system fluctuations. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a calibration method for short-term rainfall forecast and a computer-readable storage medium, which can improve the accuracy of short-term rainfall forecast.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a calibration method for short-term rainfall forecast, comprising:

[0007] performing a correlation analysis on the latest radar echo data of the forecast area at the latest time point and the historical radar echo data at each historical time point, and determining a historical time point corresponding to the historical radar echo data related to the latest radar echo data as the first historical time point;

[0008] performing a correlation analysis on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data at each first historical time point and within a preset first time period before the latest time point, and determining a first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data as the second historical time point;

[0009] Performing a correlation analysis on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data within a preset first time period before each second historical time point, and determining the second historical time point corresponding to the historical measured rainfall sequence data having the highest correlation with the latest measured rainfall sequence data as the third historical time point;

[0010] The short-term rainfall forecast data for the forecast area within the second time period preset after the latest time point is calibrated based on the measured rainfall data for the forecast area within the second time period preset after the third historical time point.

[0011] The present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned method when the program is executed by a processor.

[0012] The beneficial effects of the present invention are as follows: a triple comparison operation is performed through radar echo data at a single time point, radar echo sequence data within a period of time (i.e., multiple time points), and measured rainfall sequence data of the measuring station, i.e., the radar echo data between single time points are first compared to screen out historical records with a high similarity to the current situation, and then the radar echo sequence data between time periods are compared to verify the matching degree and narrow the scope, and then the current measured rainfall sequence data is compared with the historical measured rainfall sequence data of the measuring station to determine the historical time point with the highest correlation with the current situation, and finally the measured rainfall data after the historical time point is used to realize the calibration of the current short-term rainfall forecast data.

[0013] The present invention calibrates short-term rainfall forecast data by integrating historical measured rainfall data, effectively reducing short-term rainfall forecast errors, improving the accuracy of short-term rainfall forecasts, and better serving the needs of social and economic development. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a calibration method for short-term rainfall forecast according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0015] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following is a detailed description in conjunction with the embodiments and accompanying drawings.

[0016] See also Figure 1 , a calibration method for short-term rainfall forecast, comprising:

[0017] performing a correlation analysis on the latest radar echo data of the forecast area at the latest time point and the historical radar echo data at each historical time point, and determining a historical time point corresponding to the historical radar echo data related to the latest radar echo data as the first historical time point;

[0018] performing a correlation analysis on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data at each first historical time point and within a preset first time period before the latest time point, and determining a first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data as the second historical time point;

[0019] Performing a correlation analysis on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data within a preset first time period before each second historical time point, and determining the second historical time point corresponding to the historical measured rainfall sequence data having the highest correlation with the latest measured rainfall sequence data as the third historical time point;

[0020] The short-term rainfall forecast data for the forecast area within the second time period preset after the latest time point is calibrated based on the measured rainfall data for the forecast area within the second time period preset after the third historical time point.

[0021] From the above description, it can be seen that the beneficial effect of the present invention is that by fusing historical measured rainfall data, the short-term rainfall forecast data is calibrated, the short-term rainfall forecast error is effectively reduced, and the accuracy of the short-term rainfall forecast is improved.

[0022] Furthermore, the latest radar echo data of the latest time point in the forecast area and the historical radar echo data of each historical time point are subjected to correlation analysis to determine the historical time point corresponding to the historical radar echo data related to the latest radar echo data as the first historical time point, specifically:

[0023] Obtain radar echo data at the latest time point of the area to be forecasted as the latest radar echo data, and obtain radar echo data at each historical time point of the area to be forecasted as the historical radar echo data corresponding to each historical time point;

[0024] Calculating the Pearson correlation coefficient between the latest radar echo data and each historical radar echo data respectively;

[0025] If the Pearson correlation coefficient between the latest radar echo data and the historical radar echo data corresponding to a historical time point is greater than or equal to a preset first threshold, the historical time point is used as the first historical time point.

[0026] From the above description, it can be seen that by performing correlation analysis on the radar echo data of the latest time point and all historical time points, historical nodes that are more similar to the current situation are screened out.

[0027] Furthermore, the correlation analysis is performed on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data within a preset first time period before each first historical time point, and the first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data is determined as the second historical time point, specifically:

[0028] Acquire radar echo data of the area to be forecasted at the latest time point and within a preset first time period before the latest time point as the latest radar echo sequence data, and respectively acquire radar echo data of the area to be forecasted at each first historical time point and within a preset first time period before the latest time point as the historical radar echo sequence data corresponding to each first historical time point;

[0029] Calculating the Pearson correlation coefficient between the latest radar echo sequence data and each historical radar echo sequence data respectively;

[0030] If the Pearson correlation coefficient between the latest radar echo sequence data and the historical radar echo sequence data corresponding to a first historical time point is greater than or equal to a preset second threshold, the first historical time point is used as the second historical time point.

[0031] From the above description, it can be seen that by performing a correlation analysis on the radar echo sequence data within the latest period and the period corresponding to the previously preliminarily screened historical time points, the matching degree can be further verified and the scope can be narrowed down.

[0032] Furthermore, the correlation analysis is performed on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data within a preset first time period before each second historical time point, and the second historical time point corresponding to the historical measured rainfall sequence data with the highest correlation with the latest measured rainfall sequence data is determined as the third historical time point, specifically:

[0033] Obtain the measured rainfall data of each station in the forecast area at the latest time point and within a preset first time period before the latest time point as the latest measured rainfall series data of each station, and obtain the measured rainfall data of each station in the forecast area at each second historical time point and within a preset first time period before the latest time point as the historical measured rainfall series data of each station corresponding to each second historical time point;

[0034] Calculate the Pearson correlation coefficient between the latest measured rainfall series data of the same station and the historical measured rainfall series data corresponding to each second historical time point of the same station;

[0035] Calculate an average value based on the Pearson correlation coefficient between the latest measured rainfall sequence data of each measuring station and the historical measured rainfall sequence data corresponding to the same second historical time point of each measuring station to obtain the average correlation coefficient between the latest measured rainfall sequence data and the historical measured rainfall sequence data corresponding to the same second historical time point;

[0036] If the average correlation coefficient between the latest measured rainfall sequence data and the historical measured rainfall sequence data corresponding to a second historical time point is the highest, the second historical time point is used as the third historical time point.

[0037] From the above description, it can be seen that by performing a correlation analysis on the current measured rainfall series data and the historical measured rainfall series data of the measuring station, the historical time point with the highest correlation with the current situation is determined, which facilitates the subsequent correction of the current short-term rainfall forecast data based on the measured rainfall data after this historical time point.

[0038] Furthermore, the short-term rainfall forecast data for the forecast area within a preset second time period after the latest time point is calibrated based on the measured rainfall data for the forecast area within a preset second time period after the third historical time point, specifically as follows:

[0039] Obtain short-term rainfall forecast data for each station in the forecast area within a preset second time period after the latest time point as the latest short-term rainfall forecast sequence data for each station, and obtain respectively measured rainfall data for each station in the forecast area within a preset second time period after the third historical time point as the historical measured rainfall sequence data corresponding to each station;

[0040] Calculating the difference between the latest short-term rainfall forecast sequence data and the corresponding historical measured rainfall sequence data of the same observation station to obtain the rainfall difference sequence corresponding to the same observation station;

[0041] If each rainfall difference in the rainfall difference sequence corresponding to each station in the forecast area is less than the preset difference threshold, the latest short-term rainfall forecast sequence data and the historical measured rainfall sequence data of each station in the forecast area are output; otherwise, the historical measured rainfall sequence data and the rainfall difference sequence corresponding to each station in the forecast area are output, and the historical measured rainfall sequence data are used as the calibrated short-term rainfall forecast data.

[0042] From the above description, it can be seen that by fusing historical measured rainfall data and calibrating the short-term rainfall forecast data, the short-term rainfall forecast error can be effectively reduced and the accuracy of the short-term rainfall forecast can be improved.

[0043] The present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned method when the program is executed by a processor.

[0044] Example 1

[0045] Please refer to Figure 1 , Embodiment 1 of the present invention is: a calibration method for short-term rainfall forecast, which can be applied to rainfall forecast based on weather radar.

[0046] In this embodiment, the Pearson correlation coefficient is used to measure the degree of correlation between two sets of data. The Pearson correlation coefficient is a commonly used statistic used to measure the strength of the linear correlation between two variables. It can reflect the degree of correlation between the two variables, and its value range is between -1 and 1. If the correlation coefficient is close to 1, it indicates that there is a completely positive linear relationship between the two variables; if it is close to -1, it indicates that there is a completely negative linear relationship; if it is close to 0, it indicates that there is no linear relationship between the two variables. The calculation method of the Pearson correlation coefficient is as follows:

[0047] r=Cov(X,Y) / (σX×σY)

[0048] Where Cov(X,Y) represents the covariance of variables X and Y, σX and σY represent the standard deviations of variables X and Y, respectively. By calculating the covariance and standard deviation, we can obtain the correlation coefficient r between the two variables.

[0049] In this embodiment, when the correlation coefficient of two sets of data is closer to 1, the similarity between them is stronger.

[0050] like Figure 1 As shown, the method of this embodiment includes the following steps:

[0051] S1: performing correlation analysis on the latest radar echo data of the latest time point in the forecast area and the historical radar echo data of each historical time point, and determining a historical time point corresponding to the historical radar echo data related to the latest radar echo data as the first historical time point.

[0052] Specifically, first, the radar echo data of the latest time point of the area to be forecasted is obtained as the latest radar echo data, and the radar echo data of each historical time point of the area to be forecasted is obtained respectively as the historical radar echo data corresponding to each historical time point.

[0053] Then, the Pearson correlation coefficients between the latest radar echo data and each historical radar echo data are calculated, and the historical radar echo data whose Pearson correlation coefficients with the latest radar echo data are greater than or equal to a preset first threshold are determined. The historical time points corresponding to these historical radar echo data are regarded as the first historical time points. That is, if the Pearson correlation coefficient between the historical radar echo data corresponding to a certain historical time point and the latest radar echo data is greater than or equal to the preset first threshold, then the historical time point is regarded as the first historical time point.

[0054] In this embodiment, the first threshold is 0.9.

[0055] For example, assume that the correlation coefficients between the radar echo data at the current time point 2023.01.30 18:00:00 (i.e., 18:00 on January 30, 2023) and the radar echo data at the historical time points 2022.01.10 15:30:00 and 2022.10.10 11:35:00 are 0.97 and 0.95, respectively. Both are greater than the first threshold value. In this case, these two historical time points are taken as the first historical time points.

[0056] If the Pearson correlation coefficients between all historical radar echo data and the latest radar echo data are less than or equal to 0.9, the process ends.

[0057] Furthermore, before this step, the area to be forecasted is gridded according to the grid information of the radar echo data. In this embodiment, the area to be forecasted is divided into 2km×2km grids. When the radar echo data of the area to be forecasted is obtained, the radar echo data of all grids in the area to be forecasted is obtained.

[0058] S2: performing correlation analysis on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data at each first historical time point and within a preset first time period before the latest time point, and determining a first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data as the second historical time point.

[0059] Specifically, first, radar echo data for the forecast area at the latest time point and within a preset first time period before the latest time point is obtained as the latest radar echo sequence data. Then, radar echo data for the forecast area at each first historical time point and within a preset first time period before the latest time point is obtained as the historical radar echo sequence data corresponding to each first historical time point. In this embodiment, the first time period is 1 hour.

[0060] Then, the Pearson correlation coefficients between the latest radar echo sequence data and each historical radar echo sequence data are calculated, and historical radar echo sequence data whose Pearson correlation coefficients with the latest radar echo sequence data are greater than or equal to a preset second threshold are determined. The first historical time points corresponding to these historical radar echo sequence data are used as second historical time points. That is, if the Pearson correlation coefficient between the historical radar echo sequence data corresponding to a first historical time point and the latest radar echo sequence data is greater than or equal to the preset second threshold, then the first historical time point is used as the second historical time point.

[0061] In this embodiment, the second threshold is the same as the first threshold, which is also 0.9.

[0062] For example, the current time point is 2023.01.30 18:00:00, and the first historical time point includes 2022.01.10 15:30:00 and 2022.10.10 11:35:00. Then, the radar echo data during the period of 2023.01.30 17:00:00-2023.01.30 18:00:00 is correlated with the radar echo data during the periods of 2022.01.10 14:30:00-2022.01.10 15:30:00 and 2022.10.10 10:35:00-2022.10.10 11:35:00. Assuming that 2023.01.30 17:00:00-2023.01.30 The correlation coefficient between the radar echo data at 18:00:00 and the radar echo data from 2022.01.10 14:30:00 to 2022.01.10 15:30:00 is 0.93. Therefore, 2022.01.10 15:30:00 is used as the second historical time point.

[0063] If the Pearson correlation coefficients between all historical radar echo sequence data and the latest radar echo sequence data are less than or equal to 0.9, the process ends.

[0064] S3: Perform correlation analysis on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data at each second historical time point and within a preset first time period before the latest time point, and determine the second historical time point corresponding to the historical measured rainfall sequence data with the highest correlation with the latest measured rainfall sequence data as the third historical time point.

[0065] Specifically, first, the measured rainfall data of each measuring station in the area to be forecasted at the latest time point and within the first preset time period before that are obtained respectively as the latest measured rainfall series data of each measuring station, and the measured rainfall data of each measuring station in the area to be forecasted at each second historical time point and within the first preset time period before that are obtained respectively as the historical measured rainfall series data of each measuring station corresponding to each second historical time point.

[0066] Then, the Pearson correlation coefficients between the latest measured rainfall series data of the same station and the historical measured rainfall series data corresponding to each second historical time point of the same station are calculated respectively, thereby obtaining the Pearson correlation coefficients between the latest measured rainfall series data of each station and the historical measured rainfall series data corresponding to each second historical time point of each station.

[0067] Next, the average value of the Pearson correlation coefficient between the latest measured rainfall series data of each measuring station and the historical measured rainfall series data corresponding to the same second historical time point of each measuring station is calculated to obtain the average correlation coefficient corresponding to the second historical time point, that is, the average correlation coefficient between the latest measured rainfall series data and the historical measured rainfall series data corresponding to the second historical time point.

[0068] Finally, the average correlation coefficients corresponding to each historical time point are compared, and the second historical time point with the highest average correlation coefficient is selected as the third historical time point. That is, if the average correlation coefficient between the latest measured rainfall series data and the historical measured rainfall series data corresponding to the second historical time point is the highest, then the second historical time point is selected as the third historical time point.

[0069] S4: Calibrate the short-term rainfall forecast data of the forecast area within the second time period preset after the latest time point based on the measured rainfall data of the forecast area within the second time period preset after the third historical time point.

[0070] Specifically, first, the short-term rainfall forecast data for each station within the forecast area is obtained within a preset second time period after the latest time point, as the latest short-term rainfall forecast series data for each station. Furthermore, the measured rainfall data for each station within the forecast area is obtained within a preset second time period after the third historical time point, as the historical measured rainfall series data corresponding to each station. In this embodiment, the second time period can be 1 hour, 2 hours, or 3 hours.

[0071] Then, for each station in the forecast area, the difference between the latest short-term rainfall forecast series and the historical measured rainfall series is calculated to obtain the rainfall difference series corresponding to each station. For example, if the short-term rainfall forecast data for station A at the current time point 2023.01.30 18:00:00 is 1.4mm, and the measured rainfall data at the third historical time point 2022.01.10 15:30:00 is 1.5mm, then the first rainfall difference in the rainfall difference series for station A is 0.1mm, and so on.

[0072] If the rainfall difference in the rainfall difference sequence for all stations in the forecast area is less than a preset difference threshold, the latest short-term rainfall forecast sequence data and historical measured rainfall sequence data for each station in the forecast area are output and displayed simultaneously. Otherwise, the historical measured rainfall sequence data and rainfall difference sequence corresponding to each station in the forecast area are output to assist in manually correcting the short-term rainfall forecast data for the current period. In this embodiment, the historical measured rainfall sequence data is directly used as the calibrated short-term rainfall forecast data. By integrating historical measured rainfall data, the accuracy of the short-term rainfall forecast is improved.

[0073] Short-term rainfall forecasts must be accurate and stable. Statistical forecasting methods require statistical analysis of large amounts of historical data, resulting in high investment costs, low statistical stability, and low forecast stability. This embodiment uses a machine to replace manual data comparison and correlation analysis of radar echo data, leveraging historical meteorological patterns to assist in future forecasts and improve forecast stability. Furthermore, historical measured data can be fully integrated to calibrate short-term rainfall forecast data in situations where weather systems fluctuate significantly, mitigating the impact of geographical factors and radar data, reducing errors in weather radar extrapolation, and making the calibrated data more valuable for reference.

[0074] Example 2

[0075] This embodiment is a computer-readable storage medium corresponding to the above embodiment, on which a computer program is stored. When the program is executed by a processor, the various steps of the calibration method for short-term rainfall forecast in the above embodiment are implemented, and the same technical effects can be achieved, which will not be repeated here.

[0076] In summary, the present invention provides a calibration method and computer-readable storage medium for short-term rainfall forecasts. The method first compares radar echo data between single time points to select historical records with a high degree of similarity to the current situation. The method then compares radar echo sequence data between time periods to verify the matching degree and narrow the scope. The method then compares the current measured rainfall sequence data with the historical measured rainfall sequence data at the station to determine the historical time point with the highest correlation to the current situation. Finally, the method calibrates the current short-term rainfall forecast data using measured rainfall data after this historical time point. The present invention calibrates short-term rainfall forecast data by integrating historical measured rainfall data, effectively reducing short-term rainfall forecast errors and improving the accuracy of short-term rainfall forecasts.

[0077] The above descriptions are merely embodiments of the present invention and are not intended to limit the scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the scope of the present invention's patent protection.

Claims

1. A calibration method for short-term rainfall forecast, characterized in that: include: performing a correlation analysis on the latest radar echo data of the forecast area at the latest time point and the historical radar echo data at each historical time point, and determining a historical time point corresponding to the historical radar echo data related to the latest radar echo data as the first historical time point; performing a correlation analysis on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data at each first historical time point and within a preset first time period before the latest time point, and determining a first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data as the second historical time point; Performing a correlation analysis on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data within a preset first time period before each second historical time point, and determining the second historical time point corresponding to the historical measured rainfall sequence data having the highest correlation with the latest measured rainfall sequence data as the third historical time point; Obtain short-term rainfall forecast data for each station in the forecast area within a preset second time period after the latest time point as the latest short-term rainfall forecast sequence data for each station, and obtain respectively measured rainfall data for each station in the forecast area within a preset second time period after the third historical time point as the historical measured rainfall sequence data corresponding to each station; Calculating the difference between the latest short-term rainfall forecast sequence data and the corresponding historical measured rainfall sequence data of the same observation station to obtain the rainfall difference sequence corresponding to the same observation station; If each rainfall difference in the rainfall difference sequence corresponding to each station in the forecast area is less than the preset difference threshold, the latest short-term rainfall forecast sequence data and the historical measured rainfall sequence data of each station in the forecast area are output; otherwise, the historical measured rainfall sequence data and the rainfall difference sequence corresponding to each station in the forecast area are output, and the historical measured rainfall sequence data are used as the calibrated short-term rainfall forecast data.

2. The calibration method for short-term rainfall forecast according to claim 1, characterized in that: The correlation analysis is performed on the latest radar echo data of the latest time point in the forecast area and the historical radar echo data of each historical time point, and a historical time point corresponding to the historical radar echo data related to the latest radar echo data is determined as the first historical time point. Specifically, Obtain radar echo data at the latest time point of the area to be forecasted as the latest radar echo data, and obtain radar echo data at each historical time point of the area to be forecasted as the historical radar echo data corresponding to each historical time point; Calculating the Pearson correlation coefficient between the latest radar echo data and each historical radar echo data respectively; If the Pearson correlation coefficient between the latest radar echo data and the historical radar echo data corresponding to a historical time point is greater than or equal to a preset first threshold, the historical time point is used as the first historical time point.

3. The calibration method for short-term rainfall forecast according to claim 1, characterized in that: The method of performing correlation analysis on the latest radar echo sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical radar echo sequence data at each first historical time point and within a preset first time period before the latest time point, and determining a first historical time point corresponding to the historical radar echo sequence data related to the latest radar echo sequence data as the second historical time point, is specifically: Acquire radar echo data of the area to be forecasted at the latest time point and within a preset first time period before the latest time point as the latest radar echo sequence data, and respectively acquire radar echo data of the area to be forecasted at each first historical time point and within a preset first time period before the latest time point as the historical radar echo sequence data corresponding to each first historical time point; Calculating the Pearson correlation coefficient between the latest radar echo sequence data and each historical radar echo sequence data respectively; If the Pearson correlation coefficient between the latest radar echo sequence data and the historical radar echo sequence data corresponding to a first historical time point is greater than or equal to a preset second threshold, the first historical time point is used as the second historical time point.

4. The calibration method for short-term rainfall forecast according to claim 1, characterized in that: The method of performing correlation analysis on the latest measured rainfall sequence data of the forecast area at the latest time point and within a preset first time period before the latest time point and the historical measured rainfall sequence data at each second historical time point and within a preset first time period before the latest time point, and determining the second historical time point corresponding to the historical measured rainfall sequence data with the highest correlation with the latest measured rainfall sequence data as the third historical time point, specifically: Obtain the measured rainfall data of each station in the forecast area at the latest time point and within a preset first time period before the latest time point as the latest measured rainfall series data of each station, and obtain the measured rainfall data of each station in the forecast area at each second historical time point and within a preset first time period before the latest time point as the historical measured rainfall series data of each station corresponding to each second historical time point; Calculate the Pearson correlation coefficient between the latest measured rainfall series data of the same station and the historical measured rainfall series data of the same station corresponding to each second historical time point; Calculate an average value based on the Pearson correlation coefficient between the latest measured rainfall sequence data of each measuring station and the historical measured rainfall sequence data corresponding to the same second historical time point of each measuring station to obtain the average correlation coefficient between the latest measured rainfall sequence data and the historical measured rainfall sequence data corresponding to the same second historical time point; If the average correlation coefficient between the latest measured rainfall sequence data and the historical measured rainfall sequence data corresponding to a second historical time point is the highest, the second historical time point is used as the third historical time point.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images