A regional climate data analysis system and method based on artificial intelligence

Abstract

The present invention discloses a regional climate data analysis system and method based on artificial intelligence, which relates to the field of regional climate data analysis. Climate data is obtained through meteorological data acquisition equipment and a database; the time point when the water vapor accumulation reaches the precipitation condition is analyzed; the precipitation prediction period is obtained according to the analysis results of the climate factors affecting the probability of precipitation; climate characteristic data affecting the time of precipitation arrival is analyzed, and the precipitation prediction period is adjusted according to the analysis results. The technical solution of the present invention that determines the precipitation probability through dew point temperature analysis and then issues a rainfall prediction improves the accuracy of precipitation prediction and reduces the probability of precipitation prediction errors; and the precipitation prediction period is adjusted according to different precipitation conditions to achieve the effect of leaving sufficient prevention time for heavy rain or emergency rainstorms. The present invention adjusts the precipitation prediction period by analyzing different precipitation conditions, taking into account both the accuracy and timeliness of precipitation prediction.

Description

Technical Field

[0001] The present invention relates to the field of regional climate data analysis, and in particular to a regional climate data analysis system and method based on artificial intelligence. Background Art

[0002] Regional climate data analysis technology refers to the technical methods for collecting, processing, analyzing and interpreting climate data for specific geographical areas; these technologies are widely used in climate change research, meteorological forecasting, agricultural planning, disaster assessment and other fields; the climate system is a complex nonlinear dynamic system involving a large number of mutually coupled physical, chemical, biological and human processes; traditional empirical statistical methods are difficult to accurately describe the climate nonlinearly and predict climate transitions, and although climate prediction dynamic models can simulate some nonlinear processes, errors continue to accumulate under long-term integration, and the prediction skills for temperature, precipitation and extreme climate events are not high on the sub-seasonal to seasonal scale; the rapid development of artificial intelligence technology is profoundly changing the technical system of regional climate analysis, significantly improving data processing efficiency and model accuracy. and forecasting capabilities; scientific and reliable future projections are of great significance to climate change adaptation and the formulation of relevant policies, and multi-model ensembles are one of the effective ways to achieve this goal; there are still many problems in the current regional climate analysis process, such as the fact that precipitation in different regions will be affected by geographical characteristics, seasonal characteristics and individual special circumstances; and the analysis and forecast of precipitation is very important for people's production and life, but there are still problems with the current forecast of precipitation conditions in the region. When analyzing climate data, it is difficult to balance the accuracy and timeliness of predictions when choosing the time point for analyzing precipitation conditions. In addition, the current precipitation forecast cycle cannot be automatically adjusted according to actual conditions, and the precipitation forecast cycle model is not adaptable enough, resulting in the problem of incompatibility of the forecast model. Summary of the Invention

[0003] The purpose of the present invention is to provide a regional climate data analysis system and method based on artificial intelligence to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a regional climate data analysis method based on artificial intelligence, the method comprising the following steps:

[0005] S1. Obtain air specific humidity, whole-layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database;

[0006] S2. Based on the analysis results of the entire layer of precipitable water, analyze the time point when the water vapor accumulation reaches the precipitation conditions;

[0007] S3. Obtaining a precipitation forecast period based on the analysis results of climate factors that affect the probability of precipitation and the time point when precipitation conditions are met;

[0008] S4. Analyze the climate characteristic data that affects the time of precipitation arrival, and adjust the predicted precipitation period based on the analysis results.

[0009] Furthermore, in step S1: the area where climate data monitoring is to be performed is divided into m parts, {B1, B2, ..., B m}, real-time data collection of area B by radiosonde i The specific humidity of the air above is q, where i = 1, 2, ... m; the area B is monitored by GPS water vapor monitoring stations and satellite remote sensing PWV monitoring equipment i The total precipitation amount Q above the whole layer; and the dew point temperature T is collected by using the radiosonde dew point measurement equipment L ; Real-time collection of atmospheric vertical rise rate ω by wind profiler radar and sounding instrument; Collection of area B by weather station sensor i The average relative humidity R i .

[0010] Furthermore, in step S2, the collected climate data is analyzed to determine whether the real-time collected air specific humidity q value increases over time. The change in specific humidity over time is expressed as W and calculated using the following formula:

[0011] ;

[0012] The analysis results of W are as follows:

[0013] If W ≤ 0, it means that the specific humidity of the air remains constant or decreases over time, judging the trend of no water vapor accumulation in the air;

[0014] If W>0, it means that the specific humidity of the air increases with time, and it is judged that there is a trend of gradual accumulation of water vapor in the air, and the analysis of the precipitable water content of the entire layer is initiated;

[0015] When W>0 is detected, the area B is started i The analysis and detection of the precipitation amount of the entire layer above, and the start detection time t1 are recorded, and the critical value Q of the precipitation amount of the entire layer is set s In the process of analyzing whether the water vapor accumulation reaches the precipitation condition, the critical value of the precipitation amount in the whole layer Q s With area B i The mean relative humidity R iThe critical value of the whole layer of precipitable water is related to the numerical size of . By linking the critical value of the whole layer of precipitable water with the average relative humidity in the air, the critical value of the whole layer of precipitable water can be adjusted according to the relative humidity of the air in different areas, which can improve the accuracy of precipitation prediction. The critical value of the whole layer of precipitable water is calculated by the following formula:

[0016] ;

[0017] Among them, Q s0 Indicates the saturation value of the entire layer of precipitable water. The real-time collected precipitable water Q of the entire layer is compared with the critical value of the precipitable water of the entire layer. The analysis results are as follows:

[0018] If Q≤Q s , then it means area B i If the water vapor accumulation in the upper air is insufficient to meet the conditions for precipitation, the precipitation in the entire layer will continue to be monitored;

[0019] If Q>Q s , then it means area B i If the amount of water vapor accumulated in the sky has reached the conditions for precipitation, the time when the precipitation conditions are reached is recorded as t2.

[0020] Further, in step S3: after determining that the precipitation condition is met, the dew point temperature is analyzed, and the dew point temperature T L The trend over time is expressed as M and is calculated using the following formula:

[0021] ;

[0022] The trend of dew point temperature changing over time is analyzed, and the analysis results are as follows:

[0023] If M≤0, it means that the dew point temperature has no upward trend, and the probability of precipitation is judged to be low. Then, M is continuously analyzed until M>0, and the moment when M>0 is detected is recorded as t3. After monitoring M>0, a delayed observation time Δt0 is set. During the set delayed observation time Δt0, M is detected in real time. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high, and a warning is issued for area B. iPrecipitation forecast; if within the first delayed observation time Δt0, it is detected that M refers to the real-time monitoring value that is always greater than 0, then after the first delayed observation value ends, another delayed observation time Δt0 is superimposed, and M is continuously tested until the M value always remains greater than 0, and the number of superimposed observation time periods n is recorded, and the prediction period T1=(t3-t1)+nΔt0 of this precipitation situation is recorded; by analyzing the change trend of the dew point temperature over time, the probability of rainfall is judged, and when it is judged that the probability of rainfall is high, a precipitation forecast is issued, and the precipitation forecast period is recorded. The technical solution of judging the precipitation probability by analyzing the dew point temperature and then issuing the rainfall forecast improves the accuracy of precipitation forecast and reduces the probability of precipitation forecast errors;

[0024] If M>0, it means that the dew point temperature is on an upward trend. After monitoring M>0, set the delayed observation time Δt0. During the delayed observation time, perform real-time detection on M. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high and issue a warning for area B. i Precipitation forecast; if within the first delayed observation time Δt0, it is detected that M refers to keeping the real-time monitoring value always greater than 0, then after the end of the first delayed observation value, another delayed observation time Δt0 is superimposed, and M is continued to be detected until the M value always remains greater than 0. The number n of superimposed observation time lengths is recorded, and the prediction period T2=(t2-t1)+nΔt0 of this precipitation situation is recorded.

[0025] Furthermore, in step S4: the growth rate of the whole layer of precipitable water and the vertical rise rate of the atmosphere can affect the size of the time difference from the issuance of the precipitation forecast to the occurrence of precipitation; when monitoring the changes in the whole layer of precipitable water and atmospheric pressure, the growth rate of the whole layer of precipitable water is analyzed. , the growth rate of the whole layer of precipitable water is obtained by the following formula: ; and simultaneously analyze the vertical ascent rate of the atmosphere ω; set the growth rate threshold of the entire layer of precipitable water and atmospheric vertical ascent rate threshold The relationship between the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere and the set threshold is obtained by analyzing the relationship, and the precipitation forecast period is adjusted according to the analysis results. The analysis results are as follows:

[0026] like ≤ And ω≤ , it means that the growth rate of the whole layer of precipitable water and the vertical ascent rate of the atmosphere are moderate, and the precipitation forecast period will not be adjusted;

[0027] like > And ω≤ , it means that the growth rate of the whole layer of precipitable water is high, the vertical ascent rate of the atmosphere is moderate, and a large amount of precipitation is predicted; the high growth rate of the whole layer of precipitable water will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation. Therefore, the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>xQ s When , it is judged that the precipitation condition is met, where x represents the proportional coefficient of the reduction ratio of the critical value of precipitation in the whole layer when the growth rate of precipitation in the whole layer is high. ; That is, it is judged that the precipitation conditions are met; the time when the precipitation forecast is issued is recorded as t4, and the optimal precipitation forecast period in this case is recorded as T3=(t4-t1)+nΔt0;

[0028] like ≤ And ω> , it means that the growth rate of the whole layer of precipitable water is moderate, the vertical ascent rate of the atmosphere is high, and emergency precipitation is predicted; the high vertical ascent rate of the atmosphere will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation, so the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>yQ s When , it is judged that the precipitation condition is met, where y represents the proportional coefficient of the reduction ratio of the critical value of the precipitation of the entire layer when the vertical rising rate of the atmosphere is high. ; and directly issue a precipitation forecast; record the time of issuing the precipitation forecast as t5, and record the optimal precipitation forecast period in this case as T4=t5-t1;

[0029] like > And ω> , it means that the growth rate of the entire layer of precipitation and the vertical rise rate of atmospheric pressure are both too high, and an emergency rainstorm is predicted; when it is detected that the growth rate of the entire layer of precipitation and the vertical rise rate of atmospheric pressure are both sharply increased, a precipitation forecast is directly issued, and the time of issuing the precipitation forecast is recorded as t6, and the optimal precipitation forecast period in this case is recorded as T5=t6-t1; the analysis of the growth rate of the entire layer of precipitation and the vertical rise rate of the atmosphere achieves the effect of distinguishing different precipitation situations, and adjusts the precipitation forecast period according to different precipitation situations, so as to leave sufficient prevention time for heavy rain or emergency rainstorm. By adjusting the precipitation forecast period through analysis of different precipitation situations, both the accuracy and timeliness of precipitation forecast are taken into account.

[0030] An artificial intelligence-based regional climate data analysis system, the system includes a climate data acquisition module, a precipitation condition analysis module, a precipitation forecast period analysis module, and a precipitation forecast period adjustment module;

[0031] The climate data acquisition module obtains air specific humidity, whole layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database;

[0032] The precipitation condition analysis module is used to analyze the time point when the water vapor accumulation reaches the precipitation condition based on the analysis results of the entire layer of precipitable water;

[0033] The precipitation prediction period analysis module is used to obtain the precipitation prediction period based on the analysis results of the climate factors that affect the probability of precipitation and the time point when the precipitation conditions are met;

[0034] The precipitation prediction period adjustment module is used to analyze climate characteristic data that affects the time of precipitation arrival, and adjust the period of predicted precipitation according to the analysis results.

[0035] Furthermore, the climate data acquisition module collects the regional air specific humidity in real time through the radiosonde, monitors the whole layer of precipitable water through the GPS water vapor monitoring station and satellite remote sensing PWV monitoring equipment; and collects the dew point temperature T using the radiosonde dew point measurement equipment. L ; Real-time collection of atmospheric vertical ascent rate through wind profiler radar and sounding instrument; and collection of average relative humidity through weather station sensors.

[0036] Furthermore, the precipitation condition analysis module includes a specific humidity analysis unit and a precipitation condition analysis unit; the specific humidity analysis unit is used to analyze the change value of the collected specific humidity data over time, and determine whether water vapor accumulation has occurred in the air based on the analysis results; the precipitation condition analysis unit analyzes the accumulation amount of the entire layer of precipitable water above the analysis area to determine whether the accumulation value of the entire layer of precipitable water meets the precipitation conditions.

[0037] Furthermore, the precipitation prediction cycle analysis module is used to analyze the changing trend of the dew point temperature, determine the probability of precipitation based on the analysis results, and issue a precipitation forecast when the probability of precipitation is high, and record the obtained precipitation prediction cycle.

[0038] Furthermore, the precipitation prediction period adjustment module analyzes the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere, and adjusts the precipitation situation and the precipitation prediction period according to the analysis results.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention first analyzes the change value of the specific humidity data over time, and judges whether water vapor accumulation has occurred in the air through the change trend of the specific humidity over time, so as to determine the time node for starting precipitation prediction. After determining that water vapor has started to accumulate, the entire layer of precipitable water above the area is analyzed. After judging that the water vapor accumulation has reached the precipitation condition, the probability of rainfall is judged by analyzing the change trend of the dew point temperature over time. When it is judged that the probability of rainfall is high, a precipitation forecast is issued, and the precipitation forecast period is recorded. The technical solution of judging the precipitation probability by analyzing the dew point temperature and then issuing the rainfall forecast improves the accuracy of precipitation prediction and reduces the error of precipitation prediction. The probability of error is reduced, and when analyzing whether the water vapor accumulation reaches the precipitation conditions, the critical value of the whole layer of precipitable water is linked to the average relative humidity in the air. The critical value of the whole layer of precipitable water can be adjusted according to the relative humidity of the air in different areas, which can improve the accuracy of precipitation prediction; while monitoring the accumulation value of the whole layer of precipitation, the detection of the growth rate of the whole layer of precipitation and the vertical ascent rate of the atmosphere is started, so as to distinguish different precipitation conditions, and adjust the precipitation prediction period according to different precipitation conditions, so as to achieve the effect of leaving sufficient prevention time for heavy rain or emergency rainstorms. The present invention adjusts the precipitation prediction period by analyzing different precipitation conditions, taking into account both the accuracy and timeliness of precipitation prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of a method flow of a regional climate data analysis method based on artificial intelligence according to the present invention;

[0042] Figure 2 This is a structural diagram of an artificial intelligence-based regional climate data analysis system of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] like Figure 1-Figure 2 As shown, the present invention provides a technical solution, a regional climate data analysis method based on artificial intelligence, the method comprising the following steps:

[0045] S1. Obtain air specific humidity, whole-layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database;

[0046] S2. Based on the analysis results of the entire layer of precipitable water, analyze the time point when the water vapor accumulation reaches the precipitation conditions;

[0047] S3. Obtaining a precipitation forecast period based on the analysis results of climate factors that affect the probability of precipitation and the time point when precipitation conditions are met;

[0048] S4. Analyze the climate characteristic data that affects the time of precipitation arrival, and adjust the predicted precipitation period based on the analysis results.

[0049] In step S1: the area where climate data monitoring is to be carried out is divided into m parts, {B1, B2, ..., B m}, real-time data collection of area B by radiosonde i The specific humidity of the air above is q, where i = 1, 2, ... m; the area B is monitored by GPS water vapor monitoring stations and satellite remote sensing PWV monitoring equipment i The total precipitation amount Q above the whole layer; and the dew point temperature T is collected by using the radiosonde dew point measurement equipment L ; Real-time collection of atmospheric vertical rise rate ω by wind profiler radar and sounding instrument; Collection of area B by weather station sensor i The average relative humidity R i .

[0050] In step S2: the collected climate data is analyzed to determine whether the real-time collected air specific humidity q value increases over time. The change in specific humidity over time is expressed as W and calculated using the following formula:

[0051] ;

[0052] The analysis results of W are as follows:

[0053] If W ≤ 0, it means that the specific humidity of the air remains constant or decreases over time, judging the trend of no water vapor accumulation in the air;

[0054] If W>0, it means that the specific humidity of the air increases with time, and it is judged that there is a trend of gradual accumulation of water vapor in the air, and the analysis of the precipitable water content of the entire layer is initiated;

[0055] When W>0 is detected, the area B is started i The analysis and detection of the precipitation amount of the entire layer above, and the start detection time t1 are recorded, and the critical value Q of the precipitation amount of the entire layer is set s In the process of analyzing whether the water vapor accumulation reaches the precipitation condition, the critical value of the precipitation amount in the whole layer Q s With area B i The mean relative humidity R i The critical value of the precipitation amount in the whole layer is calculated by the following formula:

[0056] ;

[0057] Among them, Q s0 Indicates the saturation value of the entire layer of precipitable water. The real-time collected precipitable water Q of the entire layer is compared with the critical value of the precipitable water of the entire layer. The analysis results are as follows:

[0058] If Q≤Q s , then it means area B i If the water vapor accumulation in the upper air is insufficient to meet the conditions for precipitation, the precipitation in the entire layer will continue to be monitored;

[0059] If Q>Q s , then it means area B i If the amount of water vapor accumulated in the sky has reached the conditions for precipitation, the time when the precipitation conditions are reached is recorded as t2.

[0060] In step S3: after determining that the precipitation condition has been met, the dew point temperature is analyzed. L The trend over time is expressed as M and is calculated using the following formula:

[0061] ;

[0062] The trend of dew point temperature changing over time is analyzed, and the analysis results are as follows:

[0063] If M≤0, it means that the dew point temperature has no upward trend, and the probability of precipitation is judged to be low. Then, M is continuously analyzed until M>0, and the moment when M>0 is detected is recorded as t3. After monitoring M>0, a delayed observation time Δt0 is set. During the set delayed observation time Δt0, M is detected in real time. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high, and a warning is issued for area B. i Precipitation forecast; if within the first delayed observation time Δt0, it is detected that M refers to the real-time monitoring value that is always greater than 0, then after the end of the first delayed observation value, another delayed observation time Δt0 is superimposed, and M is continuously tested until the M value always remains greater than 0. The number of superimposed observation time n is recorded, and the prediction period T1=(t3-t1)+nΔt0 of this precipitation situation is recorded;

[0064] If M>0, it means that the dew point temperature is on an upward trend. After monitoring M>0, set the delayed observation time Δt0. During the delayed observation time, perform real-time detection on M. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high and issue a warning for area B. iPrecipitation forecast; if within the first delayed observation time Δt0, it is detected that M refers to keeping the real-time monitoring value always greater than 0, then after the end of the first delayed observation value, another delayed observation time Δt0 is superimposed, and M is continued to be detected until the M value always remains greater than 0. The number n of superimposed observation time lengths is recorded, and the prediction period T2=(t2-t1)+nΔt0 of this precipitation situation is recorded.

[0065] In step S4: the growth rate of the whole layer of precipitable water and the vertical rise rate of the atmosphere can affect the time difference from the issuance of precipitation forecast to the occurrence of precipitation; when monitoring the changes in the whole layer of precipitable water and atmospheric pressure, the growth rate of the whole layer of precipitable water is analyzed. , the growth rate of the whole layer of precipitable water is obtained by the following formula: ; and simultaneously analyze the vertical ascent rate of the atmosphere ω; set the growth rate threshold of the entire layer of precipitable water and atmospheric vertical ascent rate threshold The relationship between the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere and the set threshold is obtained by analyzing the relationship, and the precipitation forecast period is adjusted according to the analysis results. The analysis results are as follows:

[0066] like ≤ And ω≤ , it means that the growth rate of the whole layer of precipitable water and the vertical ascent rate of the atmosphere are moderate, and the precipitation forecast period will not be adjusted;

[0067] like > And ω≤ , it means that the growth rate of the whole layer of precipitable water is high, the vertical ascent rate of the atmosphere is moderate, and a large amount of precipitation is predicted; the high growth rate of the whole layer of precipitable water will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation. Therefore, the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>xQ s When , it is judged that the precipitation condition is met, where x represents the proportional coefficient of the reduction ratio of the critical value of precipitation in the whole layer when the growth rate of precipitation in the whole layer is high. ; Record the time when the precipitation forecast is issued as t4, and record the optimal precipitation forecast period in this case as T3=(t4-t1)+nΔt0;

[0068] like ≤ And ω> , it means that the growth rate of the whole layer of precipitable water is moderate, the vertical ascent rate of the atmosphere is high, and emergency precipitation is predicted; the high vertical ascent rate of the atmosphere will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation, so the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>yQ s When , it is judged that the precipitation condition is met, where y represents the proportional coefficient of the reduction ratio of the critical value of the precipitation of the entire layer when the vertical rising rate of the atmosphere is high. ; and directly issue a precipitation forecast; record the time of issuing the precipitation forecast as t5, and record the optimal precipitation forecast period in this case as T4=t5-t1;

[0069] like > And ω> , it means that the growth rate of the entire layer of precipitable water and the vertical rise rate of atmospheric pressure are both too high, and an emergency rainstorm is predicted; when it is detected that the growth rate of the entire layer of precipitable water and the vertical rise rate of atmospheric pressure have increased sharply, a precipitation forecast is directly issued, and the time of issuing the precipitation forecast is recorded as t6, and the optimal precipitation forecast period in this case is recorded as T5=t6-t1.

[0070] An artificial intelligence-based regional climate data analysis system, the system includes a climate data acquisition module, a precipitation condition analysis module, a precipitation forecast period analysis module, and a precipitation forecast period adjustment module;

[0071] The climate data acquisition module obtains air specific humidity, whole-layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database;

[0072] The precipitation condition analysis module is used to analyze the time point when the water vapor accumulation reaches the precipitation condition based on the analysis results of the entire layer of precipitable water;

[0073] The precipitation prediction cycle analysis module is used to obtain the precipitation prediction cycle based on the analysis results of the climate factors that affect the probability of precipitation and the time point when the precipitation conditions are met;

[0074] The precipitation forecast period adjustment module is used to analyze the climate characteristic data that affects the arrival time of precipitation, and adjust the forecast precipitation period according to the analysis results.

[0075] The climate data acquisition module collects regional air specific humidity in real time through the radiosonde, monitors the whole layer of precipitable water through GPS water vapor monitoring station and satellite remote sensing PWV monitoring equipment; uses the radiosonde dew point measurement equipment to collect the dew point temperature T L ; Real-time collection of atmospheric vertical ascent rate through wind profiler radar and sounding instrument; and collection of average relative humidity through weather station sensors.

[0076] The precipitation condition analysis module includes a specific humidity analysis unit and a precipitation condition analysis unit; the specific humidity analysis unit is used to analyze the change value of the collected specific humidity data over time, and determine whether water vapor accumulation has occurred in the air based on the analysis results; the precipitation condition analysis unit analyzes the accumulation of the entire layer of precipitable water above the analysis area to determine whether the accumulation value of the entire layer of precipitable water meets the precipitation conditions.

[0077] The precipitation forecast cycle analysis module is used to analyze the changing trend of the dew point temperature, determine the probability of precipitation based on the analysis results, and issue a precipitation forecast when the probability of precipitation is high, and record the obtained precipitation forecast cycle.

[0078] The precipitation forecast period adjustment module analyzes the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere, and adjusts the precipitation situation and precipitation forecast period according to the analysis results.

[0079] Example 1: In step S1: the area to be monitored for climate data is divided into 100 parts, {B1, B2, ..., B 100 The air humidity q over area B1 is collected in real time by radiosonde, and the water vapor monitoring station and satellite remote sensing PWV monitoring equipment are used to monitor area B. i The total precipitation amount Q above the whole layer; and the dew point temperature T is collected by using the radiosonde dew point measurement equipment L ; Real-time collection of atmospheric vertical rise rate ω by wind profiler radar and sounding instrument; Collection of area B by weather station sensor i The average relative humidity above is R1.

[0080] In step S2: the collected climate data is analyzed to determine whether the real-time collected air specific humidity q value increases over time. The change in specific humidity over time is expressed as W and calculated using the following formula:

[0081] ;

[0082] Analyze W. According to the results collected by the equipment, W=3 is obtained. If W>0, it means that the air specific humidity increases with time. It is judged that there is a trend of gradual accumulation of water vapor in the air, and the analysis of the precipitation amount of the entire layer is initiated.

[0083] When W>0 is detected, the analysis and detection of the entire layer of precipitation Q above area B1 is started, and the start detection time t1 is recorded as 8:00 on the 20th, and the critical value of the entire layer of precipitation Q is set. s In the process of analyzing whether the water vapor accumulation reaches the precipitation condition, the critical value of the precipitation amount in the whole layer Q sThe critical value of the precipitable water in the entire layer is calculated by the following formula, which is related to the value of the mean relative humidity R1=30% in area B1:

[0084] ;

[0085] Among them, Q s0 =100mm represents the saturation value of the whole layer of precipitation; the whole layer of precipitation Q collected in real time is compared with the critical value of the whole layer of precipitation, then Q s =30mm;

[0086] If Q=30mm, then Q>Q s , it means that the water vapor accumulation over area B1 has reached the conditions for precipitation, and the time when the precipitation conditions are reached is recorded as t2, which is 16:00 on the 20th.

[0087] In step S3: after determining that the precipitation condition has been met, the dew point temperature is analyzed. L The trend over time is expressed as M and is calculated using the following formula:

[0088] ;

[0089] The changing trend of the dew point temperature over time is analyzed, and the data obtained by the meteorological collection equipment is analyzed to obtain M=0.1℃ / s; if M>0, it means that the dew point temperature has an upward trend. After monitoring M>0, the delayed observation time Δt0=0.5h is set. During the delayed observation time, M is detected in real time. During the observation time, the real-time monitoring value of M is always greater than 0. It is judged that the probability of precipitation is high, and a precipitation forecast for area B1 is issued; the forecast period for recording this precipitation situation is T2=(t2-t1)+Δt0=8.5h.

[0090] In step S4: the growth rate of the whole layer of precipitable water and the vertical rise rate of the atmosphere can affect the time difference from the issuance of precipitation forecast to the occurrence of precipitation; when monitoring the changes in the whole layer of precipitable water and atmospheric pressure, the growth rate of the whole layer of precipitable water is analyzed. , the growth rate of the whole layer of precipitable water is obtained by the following formula: =3mm / h; and simultaneously analyze the vertical rise rate of the atmosphere ω=3Pa / s; set the growth rate threshold of the entire layer of precipitable water =10mm / h and atmospheric vertical ascent rate threshold =20Pa / s; by analyzing the relationship between the growth rate of the whole layer of precipitable water and the vertical rising rate of the atmosphere and the set threshold, and adjusting the precipitation forecast period according to the analysis results, then ≤ And ω≤ , it means that the growth rate of the whole layer of precipitable water and the vertical ascent rate of the atmosphere are moderate, and the precipitation forecast period will not be adjusted.

[0091] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A regional climate data analysis method based on artificial intelligence, characterized by: The method comprises the following steps: S1. Obtain air specific humidity, whole-layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database; S2. Based on the analysis results of the entire layer of precipitable water, analyze the time point when the water vapor accumulation reaches the precipitation conditions; S3. Obtaining a precipitation forecast period based on the analysis results of climate factors that affect the probability of precipitation and the time point when precipitation conditions are met; S4. Analyze climate characteristic data that affect the timing of precipitation, and adjust the predicted precipitation period based on the analysis results; In step S3: after determining that the precipitation condition has been met, the dew point temperature is analyzed. L The trend over time is expressed as M and is calculated using the following formula: ； The trend of dew point temperature changing over time is analyzed, and the analysis results are as follows: If M≤0, it means that the dew point temperature has no upward trend, and the probability of precipitation is judged to be low. Then, M is continuously analyzed until M>0, and the moment when M>0 is detected is recorded as t3. After monitoring M>0, a delayed observation time Δt0 is set. During the set delayed observation time Δt0, M is detected in real time. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high, and a warning is issued for area B. i Precipitation forecast; If it is detected that M refers to the real-time monitoring value that is always greater than 0 within the first delayed observation time Δt0, then after the first delayed observation value ends, another delayed observation time Δt0 is superimposed, and M is continuously tested until the M value always remains greater than 0. The number of superimposed observation time n is recorded, and the prediction period T1=(t3-t1)+nΔt0 of this precipitation situation is recorded; If M>0, it means that the dew point temperature is on an upward trend. After monitoring M>0, set the delayed observation time Δt0. During the delayed observation time, perform real-time detection on M. If the real-time monitoring value of M is always greater than 0 during the observation time, it is judged that the probability of precipitation is high and issue a warning for area B. i Precipitation forecast; If within the first delayed observation time Δt0, it is detected that M refers to keeping the real-time monitoring value always greater than 0, then after the end of the first delayed observation value, another delayed observation time Δt0 is superimposed, and M is continued to be detected until the M value always remains greater than 0. The number n of superimposed observation time lengths is recorded, and the prediction period T2=(t2-t1)+nΔt0 of this precipitation situation is recorded.

2. The regional climate data analysis method based on artificial intelligence according to claim 1, characterized in that: In step S1: the area where climate data monitoring is to be carried out is divided into m parts, {B1, B2, ..., B m }, real-time data collection of area B by radiosonde i The specific humidity of the air above is q, where i = 1, 2, ... m; the area B is monitored by GPS water vapor monitoring stations and satellite remote sensing PWV monitoring equipment i The total precipitation amount Q above the whole layer; and the dew point temperature T is collected by using the radiosonde dew point measurement equipment L ; Real-time collection of atmospheric vertical rise rate ω by wind profiler radar and sounding instrument; Collection of area B by weather station sensor i The average relative humidity R i .

3. The method for analyzing regional climate data based on artificial intelligence according to claim 2, characterized in that: In step S2: the collected climate data is analyzed to determine whether the real-time collected air specific humidity q value increases over time. The change in specific humidity over time is expressed as W and calculated using the following formula: ； The analysis results of W are as follows: If W ≤ 0, it means that the specific humidity of the air remains constant or decreases over time, judging the trend of no water vapor accumulation in the air; If W>0, it means that the specific humidity of the air increases with time, and it is judged that there is a trend of gradual accumulation of water vapor in the air, and the analysis of the precipitable water content of the entire layer is initiated; When W>0 is detected, the area B is started i The analysis and detection of the precipitation amount of the entire layer above, and the start detection time t1 are recorded, and the critical value Q of the precipitation amount of the entire layer is set s In the process of analyzing whether the water vapor accumulation reaches the precipitation condition, the critical value of the precipitation amount in the whole layer Q s With area B i The mean relative humidity R i The critical value of the precipitation amount in the whole layer is calculated by the following formula: ； Among them, Q s0 Indicates the saturation value of the entire layer of precipitable water. The real-time collected precipitable water Q of the entire layer is compared with the critical value of the precipitable water of the entire layer. The analysis results are as follows: If Q≤Q s , then it means area B i If the water vapor accumulation in the upper air is insufficient to meet the conditions for precipitation, the precipitation in the entire layer will continue to be monitored; If Q>Q s , then it means area B i If the amount of water vapor accumulated in the sky has reached the conditions for precipitation, the time when the precipitation conditions are reached is recorded as t2.

4. The method for analyzing regional climate data based on artificial intelligence according to claim 1, characterized in that: In step S4: the growth rate of the whole layer of precipitable water and the vertical rise rate of the atmosphere can affect the time difference from the issuance of precipitation forecast to the occurrence of precipitation; when monitoring the changes in the whole layer of precipitable water and atmospheric pressure, the growth rate of the whole layer of precipitable water is analyzed. , the growth rate of the whole layer of precipitable water is obtained by the following formula: ; and simultaneously analyze the vertical ascent rate of the atmosphere ω; set the growth rate threshold of the entire layer of precipitable water and atmospheric vertical ascent rate threshold The relationship between the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere and the set threshold is obtained by analyzing the relationship, and the precipitation forecast period is adjusted according to the analysis results. The analysis results are as follows: like ≤ And ω≤ , it means that the growth rate of the whole layer of precipitable water and the vertical ascent rate of the atmosphere are moderate, and the precipitation forecast period will not be adjusted; like > And ω≤ , it means that the growth rate of the whole layer of precipitable water is high, the vertical ascent rate of the atmosphere is moderate, and a large amount of precipitation is predicted; the high growth rate of the whole layer of precipitable water will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation. Therefore, the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>xQ s When , it is judged that the precipitation condition is met, where x represents the proportional coefficient of the reduction ratio of the critical value of precipitation in the whole layer when the growth rate of precipitation in the whole layer is high. ; Record the time when the precipitation forecast is issued as t4, and record the optimal precipitation forecast period in this case as T3=(t4-t1)+nΔt0; like ≤ And ω> , it means that the growth rate of the whole layer of precipitable water is moderate, the vertical ascent rate of the atmosphere is high, and emergency precipitation is predicted; the high vertical ascent rate of the atmosphere will shorten the time difference between the issuance of precipitation forecast and the occurrence of precipitation, so the precipitation forecast period can be shortened by adjusting the precipitation conditions for the issuance of precipitation forecast. When Q>yQ s When , it is judged that the precipitation condition is met, where y represents the proportional coefficient of the reduction ratio of the critical value of the precipitation of the entire layer when the vertical rising rate of the atmosphere is high. ; and directly issue a precipitation forecast; record the time of issuing the precipitation forecast as t5, and record the optimal precipitation forecast period in this case as T4=t5-t1; like > And ω> , it means that the growth rate of the entire layer of precipitable water and the vertical rise rate of atmospheric pressure are both too high, and an emergency rainstorm is predicted; when it is detected that the growth rate of the entire layer of precipitable water and the vertical rise rate of atmospheric pressure have increased sharply, a precipitation forecast is directly issued, and the time of issuing the precipitation forecast is recorded as t6, and the optimal precipitation forecast period in this case is recorded as T5=t6-t1.

5. An artificial intelligence-based regional climate data analysis system, characterized by: The system includes a climate data acquisition module, a precipitation condition analysis module, a precipitation forecast period analysis module, and a precipitation forecast period adjustment module; The climate data acquisition module obtains air specific humidity, whole layer precipitable water, dew point temperature and atmospheric vertical rise rate through meteorological data acquisition equipment and database; The precipitation condition analysis module is used to analyze the time point when the water vapor accumulation reaches the precipitation condition based on the analysis results of the entire layer of precipitable water; The precipitation prediction period analysis module is used to obtain the precipitation prediction period based on the analysis results of the climate factors that affect the probability of precipitation and the time point when the precipitation conditions are met; The precipitation prediction period adjustment module is used to analyze climate characteristic data that affects the time of precipitation arrival, and adjust the period of predicted precipitation according to the analysis results.

6. The artificial intelligence-based regional climate data analysis system according to claim 5, characterized in that: The climate data acquisition module collects regional air specific humidity in real time through the radiosonde, monitors the whole layer of precipitable water through the GPS water vapor monitoring station and satellite remote sensing PWV monitoring equipment; uses the radiosonde dew point measurement equipment to collect the dew point temperature T L ; Real-time collection of atmospheric vertical ascent rate through wind profiler radar and sounding instrument; and collection of average relative humidity through weather station sensors.

7. The regional climate data analysis system based on artificial intelligence according to claim 5, characterized in that: The precipitation condition analysis module includes a specific humidity analysis unit and a precipitation condition analysis unit; The specific humidity analysis unit is used to analyze the change in the collected specific humidity data over time, and determine whether water vapor accumulation has occurred in the air based on the analysis results; the precipitation condition analysis unit analyzes the accumulation of the entire layer of precipitable water above the analysis area to determine whether the accumulation value of the entire layer of precipitable water meets the precipitation conditions.

8. The artificial intelligence-based regional climate data analysis system according to claim 5, characterized in that: The precipitation prediction cycle analysis module is used to analyze the changing trend of the dew point temperature, determine the probability of precipitation based on the analysis results, and issue a precipitation forecast when the probability of precipitation is high, and record the obtained precipitation prediction cycle.

9. The regional climate data analysis system based on artificial intelligence according to claim 5, characterized in that: The precipitation prediction period adjustment module analyzes the growth rate of the entire layer of precipitable water and the vertical ascent rate of the atmosphere, and adjusts the precipitation situation and the precipitation prediction period according to the analysis results.

Images