Downscaling method and device for time scale of rainfall data, equipment and medium

Abstract

The invention discloses a downscaling method, device and equipment for a time scale of rainfall data and a medium, which are applied to the technical field of rainfall data processing. In the method, to-be-processed rainfall data of a target station included in a to-be-processed rainfall random event set is acquired, and simulated rainfall events included in the to-be-processed rainfall data are divided into simulated heavy rainfall events and simulated non-heavy rainfall events. Different downscaling methods are adopted for different types of simulated rainfall events. And for the simulated heavy rainfall event, downscaling the rainfall data of the simulated heavy rainfall event by using the historical heavy rainfall event hour rainfall pattern matched with the target historical heavy rainfall event. And for a non-heavy rainfall event, performing downscaling processing by using the target random hourly rainfall pattern corresponding to the first day order. And multiple hourly rainfall patterns are used, so that the randomness of the generated hourly-scale rainfall pattern is improved, and the flood disaster assessment effect is improved.

Description

Technical Field

[0001] The present application relates to the technical field of precipitation data processing, and in particular to a method, device, equipment and medium for downscaling the time scale of precipitation data. Background Art

[0002] The precipitation random event set is generated by weather generator, empirical orthogonal function analysis method and numerical weather model. Due to the limitations of basic data, the accuracy of the precipitation random event set generation model, the amount of calculation, etc., the generated precipitation random event set usually includes daily precipitation data. Daily precipitation data refers to precipitation data with daily data scale.

[0003] Precipitation random event sets are often used to assess precipitation conditions. For example, precipitation random event sets are used as input data for hydrological and hydraulic models to assess flood disasters. The scale of input data for hydrological and hydraulic models is usually hourly. It is necessary to downscale the daily precipitation data included in the precipitation random event set to convert the daily precipitation data into hourly precipitation data.

[0004] At present, the method of downscaling daily precipitation data lacks randomness, resulting in poor randomness of the obtained hourly precipitation data, which reduces the simulation effect of the hydrological and hydraulic models and makes it difficult to meet the needs of flood disaster assessment. Summary of the invention

[0005] In view of this, the present application provides a method, device, equipment and medium for downscaling the time scale of precipitation data, which can improve the randomness of precipitation data on an hourly scale.

[0006] The technical solutions provided by this application are as follows:

[0007] In a first aspect, the present application provides a method for downscaling the time scale of precipitation data, the method comprising:

[0008] Acquire precipitation data to be processed of a target site included in the precipitation random event set to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events;

[0009] In the precipitation data to be processed, the simulated precipitation events that meet the heavy precipitation event conditions are determined as simulated heavy precipitation events, and the simulated precipitation events other than the simulated heavy precipitation events are determined as simulated non-heavy precipitation events;

[0010] Determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly-scale precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event, wherein the historical heavy precipitation event hourly rainfall type includes an hourly precipitation ratio, and the hourly precipitation ratio is a ratio of each hour's precipitation to the precipitation on that day;

[0011] If the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy rainfall event is determined, the precipitation data of the first precipitation day included in the simulated non-heavy rainfall event is divided into hourly precipitation data according to the target random hourly rainfall type; the target random hourly rainfall type of the first day sequence is determined according to the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the target random hourly rainfall type includes the hourly precipitation ratio.

[0012] In a possible implementation, determining a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition includes:

[0013] Extracting a first characteristic value of the simulated heavy rainfall event from the precipitation data of the simulated heavy rainfall event;

[0014] Determining a matching degree between the simulated heavy precipitation event and the historical heavy precipitation event according to the first characteristic value of the simulated heavy precipitation event and the second characteristic value of the historical heavy precipitation event, wherein the historical heavy precipitation event is a historical precipitation event of the target station that meets the heavy precipitation event condition;

[0015] The historical heavy precipitation events whose matching degree meets the matching condition are taken as the target historical heavy precipitation events.

[0016] In a possible implementation, the historical heavy rainfall event belongs to a historical heavy rainfall event database, and the historical heavy rainfall event database is constructed in the following manner:

[0017] Acquire historical precipitation data of a target site, wherein the historical precipitation data includes precipitation data of a plurality of historical precipitation events;

[0018] In the historical precipitation data, historical precipitation events that meet the heavy precipitation event conditions are determined as historical heavy precipitation events;

[0019] Extracting a second characteristic value of the historical heavy rainfall event;

[0020] The historical heavy rainfall event database is constructed using the second characteristic value of the historical heavy rainfall event.

[0021] In a possible implementation, dividing the precipitation data of the simulated heavy precipitation event into hourly-scale precipitation data according to the hourly rainfall type of the historical heavy precipitation event of the target historical heavy precipitation event includes:

[0022] If the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event, then the precipitation days included in the simulated heavy precipitation event are matched one by one with the precipitation days included in the target historical heavy precipitation event in date order;

[0023] According to the hourly rainfall type of historical heavy rainfall events, including the hourly rainfall type of each rainfall day, the precipitation data of the simulated heavy rainfall event matched with the rainfall day is divided to obtain the hourly precipitation data of each rainfall day of the simulated heavy rainfall event;

[0024] If the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, then determine the matching relationship between the day with the maximum precipitation in the simulated heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event, and determine the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event according to the relative order of dates, wherein the relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation;

[0025] According to the hourly rainfall type of the precipitation day included in the hourly rainfall type of the historical heavy rainfall event, the precipitation data of the simulated heavy rainfall event matched with the precipitation day is divided to obtain the hourly-scale precipitation data of the simulated heavy rainfall event.

[0026] In a possible implementation, the method further includes:

[0027] If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and based on the second day sequence of the unmatched precipitation days, the target random hourly rain type corresponding to the second day sequence is determined, the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event are divided into hourly scale precipitation data according to the target random hourly rain type corresponding to the second day sequence.

[0028] In a possible implementation, the method further includes:

[0029] If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and there is no target random hourly rain type corresponding to the second day sequence of the unmatched precipitation days, then the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event are divided into hourly-scale precipitation data using a standard hourly rain type, and the standard hourly rain type is determined based on the precipitation data of historical precipitation events at the target site, and the standard hourly rain type includes an hourly precipitation ratio.

[0030] In a possible implementation, determining the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event includes:

[0031] According to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, query the random hourly rain type corresponding to the first day sequence, wherein the random hourly rain type corresponding to the first day sequence includes the random hourly rain type of at least one precipitation day within the day sequence range;

[0032] If the random hourly rain pattern corresponding to the first date sequence is found, the random hourly rain pattern of any precipitation day is selected from the random hourly rain patterns corresponding to the first date sequence as the target random hourly rain pattern.

[0033] In a possible implementation, the random hourly rainfall pattern belongs to a daily sequential random hourly rainfall pattern library, and the daily sequential random hourly rainfall pattern library is constructed in the following manner:

[0034] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0035] From the historical precipitation data, obtain the hourly rainfall pattern of the precipitation day within the range of each day sequence, and obtain the random hourly rainfall pattern of the day sequence;

[0036] The random hourly rainfall pattern library of each day is constructed by using the random hourly rainfall pattern of each day.

[0037] In a possible implementation, the method further includes:

[0038] If there is no target random hourly rainfall pattern corresponding to the first day sequence, the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event is divided into hourly precipitation data using a standard hourly rainfall pattern, wherein the standard hourly rainfall pattern is determined based on the precipitation data of historical precipitation events at the target station, and the standard hourly rainfall pattern includes hourly precipitation ratios.

[0039] In a possible implementation, the standard hourly rainfall type belongs to a standard hourly rainfall type library, and the standard hourly rainfall type library is constructed in the following manner:

[0040] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0041] According to the historical precipitation data, determine the average value of the precipitation ratio of the target hour in the hourly rainfall pattern of each precipitation day as the precipitation ratio of the target hour in the standard hourly rainfall pattern, where the target hour is each hour of the day;

[0042] The standard hourly rainfall pattern library is constructed using the standard hourly rainfall pattern.

[0043] In a possible implementation, the method further includes:

[0044] Acquire downscaled precipitation data of a plurality of stations belonging to the same region included in the precipitation random event set, wherein the downscaled precipitation data includes hourly-scale precipitation data of a plurality of simulated precipitation events;

[0045] From the downscaled precipitation data of the multiple stations, obtaining precipitation data of precipitation days in the same date sequence, and obtaining precipitation data of multiple stations to be reconstructed;

[0046] The order of the hourly precipitation ratios of the precipitation data of the station to be reconstructed with the largest daily precipitation is taken as the standard order;

[0047] The order of the hourly precipitation ratios in the precipitation data of other stations to be reconstructed is adjusted according to the standard order to obtain the reconstructed station precipitation data.

[0048] In a second aspect, the present application provides a device for downscaling the time scale of precipitation data, the device comprising:

[0049] An acquisition unit, used for acquiring precipitation data to be processed of a target station included in the precipitation random event set to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events;

[0050] a type determination unit, configured to determine, in the precipitation data to be processed, a simulated precipitation event that meets the heavy precipitation event condition as a simulated heavy precipitation event, and to determine a simulated precipitation event other than the simulated heavy precipitation event as a simulated non-heavy precipitation event;

[0051] A first downscaling unit is used to determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly-scaled precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event, wherein the historical heavy precipitation event hourly rainfall type includes an hourly precipitation ratio, and the hourly precipitation ratio is a ratio of each hour's precipitation to the precipitation on that day;

[0052] The second downscaling unit is used to determine the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, and divide the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event into hourly-scaled precipitation data according to the target random hourly rainfall type; the target random hourly rainfall type of the first day sequence is determined according to the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the target random hourly rainfall type includes the hourly precipitation ratio.

[0053] In a possible implementation, the first downscaling unit is used to determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, including:

[0054] The first downscaling unit is used to extract a first eigenvalue of the simulated heavy precipitation event from the precipitation data of the simulated heavy precipitation event; determine a matching degree between the simulated heavy precipitation event and the historical heavy precipitation event according to the first eigenvalue of the simulated heavy precipitation event and a second eigenvalue of the historical heavy precipitation event, wherein the historical heavy precipitation event is a historical precipitation event of the target station that meets the heavy precipitation event condition; and use the historical heavy precipitation event whose matching degree meets the matching condition as the target historical heavy precipitation event.

[0055] In a possible implementation, the historical heavy rainfall event belongs to a historical heavy rainfall event database, and the historical heavy rainfall event database is constructed in the following manner:

[0056] Acquire historical precipitation data of a target site, wherein the historical precipitation data includes precipitation data of a plurality of historical precipitation events;

[0057] In the historical precipitation data, historical precipitation events that meet the heavy precipitation event conditions are determined as historical heavy precipitation events;

[0058] Extracting a second characteristic value of the historical heavy rainfall event;

[0059] The historical heavy rainfall event database is constructed using the second characteristic value of the historical heavy rainfall event.

[0060] In a possible implementation, the first downscaling unit is used to divide the precipitation data of the simulated heavy precipitation event into hourly-scaled precipitation data according to the hourly rainfall type of the historical heavy precipitation event of the target historical heavy precipitation event, including:

[0061] The first downscaling unit is used to match the precipitation days included in the simulated heavy precipitation event with the precipitation days included in the target historical heavy precipitation event in date order if the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event; divide the precipitation data of the simulated heavy precipitation event matched with the precipitation day according to the hourly rainfall type of each precipitation day including the hourly rainfall type of the historical heavy precipitation event, and obtain the hourly scale precipitation data of each precipitation day of the simulated heavy precipitation event; if the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, determine the simulated heavy precipitation event. The matching relationship between the day with the maximum precipitation in the heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event is determined, and according to the relative order of dates, the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event is determined, and the relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation; according to the hourly rainfall type of the precipitation day included in the hourly rainfall type of the historical heavy precipitation event, the precipitation data of the simulated heavy precipitation event matched by the precipitation day is divided to obtain the hourly scale precipitation data of the simulated heavy precipitation event.

[0062] In a possible implementation manner, the device further includes:

[0063] The third downscaling unit is used for, if the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and according to the second day sequence of the unmatched precipitation days, determining the target random hourly rainfall type corresponding to the second day sequence, dividing the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event into hourly-scaled precipitation data according to the target random hourly rainfall type corresponding to the second day sequence.

[0064] In a possible implementation manner, the device further includes:

[0065] The fourth downscaling unit is used to divide the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event into hourly-scale precipitation data using a standard hourly rain type if the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and there is no target random hourly rain type corresponding to the second day sequence of the unmatched precipitation days. The standard hourly rain type is determined based on the precipitation data of historical precipitation events at the target site, and the standard hourly rain type includes an hourly precipitation ratio.

[0066] In a possible implementation, the second downscaling unit is configured to determine, according to a first day sequence of a first precipitation day included in the simulated non-heavy precipitation event, a target random hourly rainfall type corresponding to the first day sequence, including:

[0067] The second downscaling unit is used to query the random hourly rain pattern corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, wherein the random hourly rain pattern corresponding to the first day sequence includes the random hourly rain pattern of at least one precipitation day within the day sequence range; if the random hourly rain pattern corresponding to the first day sequence is queried, the random hourly rain pattern of any precipitation day is selected from the random hourly rain patterns corresponding to the first day sequence as the target random hourly rain pattern.

[0068] In a possible implementation, the random hourly rainfall pattern belongs to a daily sequential random hourly rainfall pattern library, and the daily sequential random hourly rainfall pattern library is constructed in the following manner:

[0069] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0070] From the historical precipitation data, obtain the hourly rainfall pattern of the precipitation day within the range of each day sequence, and obtain the random hourly rainfall pattern of the day sequence;

[0071] The random hourly rainfall pattern library of each day is constructed by using the random hourly rainfall pattern of each day.

[0072] In a possible implementation manner, the device further includes:

[0073] The fifth downscaling unit is used to divide the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event into hourly-scale precipitation data by using a standard hourly rain type if there is no target random hourly rain type corresponding to the first day sequence, wherein the standard hourly rain type is determined based on the precipitation data of the historical precipitation events of the target station, and the first standard hourly rain type includes an hourly precipitation ratio.

[0074] In a possible implementation, the standard hourly rainfall type belongs to a standard hourly rainfall type library, and the standard hourly rainfall type library is constructed in the following manner:

[0075] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0076] According to the historical precipitation data, determine the average value of the precipitation ratio of the target hour in the hourly rainfall pattern of each precipitation day as the precipitation ratio of the target hour in the standard hourly rainfall pattern, where the target hour is each hour of the day;

[0077] The standard hourly rainfall pattern library is constructed using the standard hourly rainfall pattern.

[0078] In a possible implementation manner, the device further includes:

[0079] A spatial reconstruction unit is used to obtain downscaled precipitation data of multiple stations belonging to the same area included in a precipitation random event set, wherein the downscaled precipitation data include hourly precipitation data of multiple simulated precipitation events; obtain precipitation data of precipitation days in the same day sequence from the downscaled precipitation data of the multiple stations to obtain precipitation data of multiple stations to be reconstructed; take the order of the hourly precipitation ratio of the precipitation data of the station to be reconstructed with the largest daily precipitation as the standard order; adjust the order of the hourly precipitation ratio in the precipitation data of other stations to be reconstructed according to the standard order to obtain the reconstructed station precipitation data.

[0080] In a third aspect, the present application provides a device, including: a processor, a memory, and a system bus;

[0081] The processor and the memory are connected via the system bus;

[0082] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes the method described in any embodiment of the first aspect.

[0083] In a fourth aspect, the present application provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium. When the instructions are executed on a terminal device, the terminal device executes the method described in any implementation of the first aspect above.

[0084] It can be seen that this application has the following beneficial effects:

[0085] The present application provides a method, device, equipment and medium for downscaling the time scale of precipitation data, obtains the precipitation data to be processed of the target site included in the set of precipitation random events to be processed, and divides the simulated precipitation events included in the precipitation data to be processed into simulated heavy precipitation events and simulated non-heavy precipitation events. Different types of simulated precipitation events use different downscaling methods. For simulated heavy precipitation events, the hourly rain type of historical heavy precipitation events of the matching target historical heavy precipitation events is used to downscale the precipitation data of the simulated heavy precipitation events. In this way, the hourly rain type of historical heavy precipitation events with similar characteristics can be used for downscaling, and the hourly rain type used for downscaling is richer, which enhances the randomness and flexibility of the downscaling process, and the precipitation data obtained at the hourly scale is more effective, which is conducive to a more accurate assessment of flood disasters. For simulated non-heavy precipitation events, the target random hourly rain type corresponding to the first day sequence is used for downscaling. The target random hourly rainfall pattern conforms to the characteristics of simulating non-heavy precipitation events, improves the randomness of the generated hourly precipitation pattern, and also improves the effectiveness of the generated hourly precipitation pattern, ensuring the comprehensiveness of the precipitation event set for flood disaster prediction and improving the effect of assessing flood disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 A schematic flow chart of a method for downscaling the time scale of precipitation data provided in an embodiment of the present application;

[0087] Figure 2 A schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application;

[0088] Figure 3 A schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application;

[0089] Figure 4 A schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application;

[0090] Figure 5 A schematic flow chart of a method for downscaling the time scale of precipitation data is provided for an embodiment of the present application;

[0091] Figure 6 A schematic diagram of the structure of a device for downscaling the time scale of precipitation data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0092] In order to facilitate the understanding and explanation of the technical solutions provided by the embodiments of the present application, the background technology of the present application will be described below.

[0093] Flood disasters are one of the most serious natural disasters. Using hydrological and hydraulic models to simulate flood processes is an effective means to prevent flood disasters. Hydrological and hydraulic models require hourly precipitation data as input data. The precipitation random event set usually constructed includes daily precipitation data. When using the precipitation random event set to simulate flood processes, it is necessary to downscale the daily precipitation data to obtain hourly precipitation data.

[0094] At present, a statistical method is usually used for precipitation data of a single station to obtain a rainstorm type. The rainstorm type is used to downscale the daily precipitation data included in the precipitation random event set. The rainstorm type is relatively fixed, and it is difficult to consider the temporal correlation and spatial correlation of precipitation under different precipitation conditions. The hourly precipitation data generated by the rainstorm type has low randomness. The simulation results of flood disasters obtained after processing by the hydrological and hydraulic model have poor randomness and are not accurate enough to meet the needs of simulating flood disasters.

[0095] Based on this, the embodiment of the present application provides a downscaling method for the time scale of precipitation data, obtains the precipitation data to be processed of the target site included in the set of precipitation random events to be processed, and divides the simulated precipitation events included in the precipitation data to be processed into simulated heavy precipitation events and simulated non-heavy precipitation events. Different types of simulated precipitation events use different downscaling methods. For simulated heavy precipitation events, the hourly rain type of historical heavy precipitation events of the matching target historical heavy precipitation events is used to downscale the precipitation data of the simulated heavy precipitation events. In this way, the hourly rain type of historical heavy precipitation events with similar characteristics can be used for downscaling, and the hourly rain type used for downscaling is richer, which enhances the randomness and flexibility of the downscaling process, and the precipitation data obtained at the hourly scale is more effective, which is conducive to a more accurate assessment of flood disasters. For simulated non-heavy precipitation events, the target random hourly rain type corresponding to the first day sequence is used for downscaling. The target random hourly rainfall pattern conforms to the characteristics of simulating non-heavy precipitation events, improves the randomness of the generated hourly precipitation pattern, and also improves the effectiveness of the generated hourly precipitation pattern, ensuring the comprehensiveness of the precipitation event set for assessing flood disasters and improving the effect of assessing flood disasters.

[0096] The downscaling method using the time scale of precipitation data provided in the embodiment of the present application can be applied to electronic devices that process precipitation data. An electronic device refers to a device with data processing capabilities, such as a server or a terminal. Among them, the terminal includes but is not limited to a smart phone, a tablet computer, a laptop computer, a personal digital assistant (PDA) or a smart wearable device. The server can be a cloud server, such as a central server in a central computing cluster, or an edge server in an edge computing cluster. Of course, the server can also be a server in a local data center. A local data center refers to a data center directly controlled by a user.

[0097] Those skilled in the art will appreciate that the above application scenario is only an example in which the implementation of the present application can be implemented. The scope of application of the implementation of the present application is not limited in any aspect of the framework.

[0098] To facilitate understanding of the technical solution provided in the embodiment of the present application, a method for downscaling the time scale of precipitation data provided in the embodiment of the present application is described below with reference to the accompanying drawings.

[0099] See also Figure 1 As shown in FIG. 1 , this figure is a flow chart of a method for downscaling the time scale of precipitation data provided in an embodiment of the present application. Figure 1 As shown, a method for downscaling the time scale of precipitation data provided in an embodiment of the present application includes S101-S104.

[0100] S101: Obtain precipitation data to be processed of a target station included in a set of precipitation random events to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events.

[0101] The set of precipitation random events to be processed is a set of precipitation random events generated in advance. The set of precipitation random events to be processed includes precipitation data to be processed of the target site. The embodiment of the present application does not limit the number of sites involved in the set of precipitation random events to be processed. The set of precipitation random events to be processed may include precipitation data to be processed of one or more sites in a region, or may include precipitation data to be processed of one or more sites in multiple regions. Among them, the region is, for example, a region divided according to a river basin. The region is, for example, a region divided according to administrative planning, such as a region obtained by dividing the city. The region can be set according to the need to process precipitation data, and the embodiment of the present application does not limit this. The target site is any one of the sites involved in the set of precipitation random events to be processed.

[0102] The precipitation data to be processed are the precipitation data of the stations that need to be temporally downscaled.

[0103] The precipitation data to be processed includes precipitation data of multiple simulated precipitation events. The embodiment of the present application does not limit the time length involved in the precipitation data to be processed, for example, including multiple simulated precipitation events under 1000 or 10000 random scenarios. The precipitation data to be processed is precipitation data on a daily scale.

[0104] The embodiment of the present application does not limit the triggering conditions for obtaining the precipitation data to be processed. As an example, in response to using the precipitation data to be processed included in the precipitation random event set to be processed as the input data of the hydrological and hydraulic model, the precipitation data to be processed included in the precipitation random event set to be processed is obtained, and the subsequent time scale downscaling steps are performed. As another example, in response to obtaining the generated precipitation random event set to be processed, the precipitation data to be processed included in the precipitation random event set to be processed is obtained, and the subsequent time scale downscaling steps are performed, so as to facilitate the subsequent direct use of the downscaled precipitation data for flood disaster simulation.

[0105] S102: In the precipitation data to be processed, the simulated precipitation events that meet the heavy precipitation event conditions are determined as simulated heavy precipitation events, and the simulated precipitation events other than the simulated heavy precipitation events are determined as simulated non-heavy precipitation events.

[0106] Among the multiple simulated precipitation events included in the precipitation data to be processed, the precipitation intensities of different simulated precipitation events may be different, and the corresponding downscaling methods are also different. First, the simulated precipitation events are divided according to the precipitation intensities of the simulated precipitation events.

[0107] The heavy precipitation event condition is used to determine the simulated heavy precipitation event. As an example, the heavy precipitation event condition is that the accumulated precipitation for any two consecutive days is greater than the precipitation threshold.

[0108] The precipitation threshold can be determined based on the historical precipitation statistical characteristics and percentile values ​​of the site. For example, the precipitation threshold is a% of the precipitation in the historical precipitation statistical characteristics of the site. For another example, the precipitation threshold is determined based on the precipitation in the historical precipitation statistical characteristics of the site and the flood precipitation characteristics of the area to which the site belongs. In this way, the precipitation threshold can be determined based on the precipitation conditions in the area to which the site belongs. For example, in areas with a relatively dry climate, the precipitation threshold can be 10 mm. In areas with a relatively humid climate, the precipitation threshold can be 50 mm.

[0109] The heavy precipitation event condition is used to determine the simulated heavy precipitation event and the simulated non-heavy precipitation event among the multiple simulated precipitation events included in the precipitation data to be processed.

[0110] The simulated precipitation events that meet the conditions of heavy precipitation events are determined as simulated heavy precipitation events. Other simulated precipitation events except simulated heavy precipitation events are determined as simulated non-heavy precipitation events.

[0111] In addition, the start date and end date of the simulated heavy precipitation event are determined. Of the two consecutive days when the accumulated precipitation is greater than the precipitation threshold, the first day when the precipitation is not 0 is the start date of this simulated heavy precipitation event. After the start date of the simulated heavy precipitation event, if the accumulated precipitation is less than the precipitation threshold for any two consecutive days, it is determined that this simulated heavy precipitation event ends. The first day of the two consecutive days when the accumulated precipitation is less than the precipitation threshold is the end date of the simulated heavy precipitation event.

[0112] After determining the simulated heavy precipitation events and the simulated non-heavy precipitation events, time scale downscaling is performed respectively according to their different characteristics.

[0113] If time scale downscaling is performed for the simulated heavy rainfall event, S103 is executed.

[0114] If time scale downscaling is performed for simulating non-heavy precipitation events, S104 is executed.

[0115] S103: Determine a target historical heavy precipitation event of a target site whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly scale precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event.

[0116] The downscaling of the time scale of the simulated heavy precipitation event needs to refer to the historical heavy precipitation event. The target historical heavy precipitation event of the target station is the historical heavy precipitation event of the target station whose matching degree with the simulated heavy precipitation event satisfies the matching degree condition. The matching degree condition is used, for example, to determine the historical heavy precipitation event with the highest matching degree with the simulated heavy precipitation event.

[0117] Historical heavy precipitation events can be judged according to heavy precipitation event conditions. As an example, a historical heavy precipitation event library is pre-constructed. The historical heavy precipitation event library includes historical heavy precipitation events. The embodiment of the present application provides a possible method for constructing a historical heavy precipitation event library, see A1-A4.

[0118] A1: Obtain historical precipitation data of the target site, where the historical precipitation data includes precipitation data of multiple historical precipitation events.

[0119] A2: In the historical precipitation data, historical precipitation events that meet the conditions for heavy precipitation events are identified as historical heavy precipitation events.

[0120] As an example, the condition for a heavy precipitation event is that the accumulated precipitation for any two consecutive days is greater than the precipitation threshold. Of the two days when the accumulated precipitation for any two consecutive days is greater than the precipitation threshold, the first day when the precipitation is not 0 is the start date of this historical heavy precipitation event. After the start date of the historical heavy precipitation event, if the accumulated precipitation for any two consecutive days is less than the precipitation threshold, it is determined that this historical heavy precipitation event has ended. The first day of the two days when the accumulated precipitation for any two consecutive days is less than the precipitation threshold is the end date of the historical heavy precipitation event.

[0121] A3: Extract the second eigenvalue of historical heavy rainfall events.

[0122] The second characteristic value of the historical heavy precipitation event includes, for example, one or more of the order of the maximum precipitation day, the daily precipitation on the maximum precipitation day, the cumulative precipitation of the precipitation event, the number of days from the start date of the precipitation event to the maximum precipitation day, and the number of days from the maximum precipitation day to the end date of the event. Among them, the maximum precipitation day is the day with the maximum precipitation.

[0123] A4: Use the second eigenvalue of historical heavy rainfall events to build a historical heavy rainfall event database.

[0124] It should be noted that A1-A4 is a method for constructing a historical heavy rainfall event library for a target site. The historical heavy rainfall event library may include the second characteristic values ​​of historical heavy rainfall events of multiple sites. The second characteristic value of the historical heavy rainfall event of each site can be obtained by adopting the steps of A1-A4.

[0125] The embodiment of the present application does not limit the way to determine the degree of matching. As an example, the first characteristic value of the simulated heavy precipitation event is extracted from the precipitation data of the simulated heavy precipitation event. The first characteristic value, for example, includes one or more of the day sequence of the maximum precipitation day, the daily precipitation on the maximum precipitation day, the cumulative precipitation of the precipitation event, the number of days of the precipitation event, the number of days from the event start date to the maximum precipitation day, and the number of days from the maximum precipitation day to the event end date. The day sequence is the order of the date in the dates included in a year. For example, the day sequence of a date is 50, which means that the date is the 50th day of the year.

[0126] The historical heavy precipitation event has a second characteristic value. The second characteristic value of the historical heavy precipitation event is extracted from the precipitation data of the historical heavy precipitation event. The second characteristic value of the historical heavy precipitation event, for example, includes one or more of the day sequence of the maximum precipitation day, the daily precipitation on the maximum precipitation day, the cumulative precipitation of the precipitation event, the number of days of the precipitation event from the start date of the event to the maximum precipitation day, and the number of days from the maximum precipitation day to the end date of the event.

[0127] The matching degree between the simulated heavy precipitation event and the historical heavy precipitation event is determined according to the first characteristic value of the simulated heavy precipitation event and the second characteristic value of the historical heavy precipitation event.

[0128] It should be noted that in some possible implementations, in order to improve the time correlation, historical heavy precipitation events that meet the time range conditions are used to calculate the matching degree. The time range conditions are, for example: the date sequence of the maximum precipitation day of the historical heavy precipitation event is within [dates-15, dates+15]. dates is the date sequence of the maximum precipitation day of the simulated heavy precipitation event. In this way, the historical heavy precipitation events can be screened from the perspective of date sequence, and the historical heavy precipitation events obtained are closer to the simulated heavy precipitation events in the time dimension, which helps to find a more matching target historical heavy precipitation event.

[0129] The embodiment of the present application does not limit the method for calculating the matching degree. The matching degree condition can be specifically determined based on the matching degree calculation method and the need for downscaling.

[0130] As an example, the matching degree is the Euclidean distance. The smaller the value of the Euclidean distance is, the higher the matching degree is. The matching degree condition is, for example, the minimum matching degree.

[0131] Specifically, the different dimensions of different eigenvalues ​​need to be considered during calculation. When calculating the Euclidean distance, the first eigenvalue of the simulated heavy rainfall event and the second eigenvalue of the historical heavy rainfall event are first standardized. The standardized formula is shown in formula (1):

[0132] Xs= (X-X min ) / ( X max -Xmin ) (1)

[0133] Among them, X is the eigenvalue of the same type in the first eigenvalue and the second eigenvalue. Xs is the eigenvalue obtained after standardization. Xmax and ΔXmin are the maximum and minimum values ​​of the eigenvalues ​​of each type in the historical heavy rainfall events of the target station, respectively. After standardization, the calculation formula of the matching degree, that is, the Euclidean distance, is:

[0134]

[0135] Among them, pmax1, psum1, day1, da1 and db1 are respectively the daily precipitation on the maximum precipitation day in the first eigenvalue after standardization, the cumulative precipitation of the precipitation event, the number of days of the precipitation event, the number of days from the event start date to the maximum precipitation day, and the number of days from the maximum precipitation day to the event end date. pmax2, psum2, day2, da2 and db2 are respectively the daily precipitation on the maximum precipitation day in the second eigenvalue after standardization, the cumulative precipitation of the precipitation event, the number of days of the precipitation event, the number of days from the event start date to the maximum precipitation day, and the number of days from the maximum precipitation day to the event end date. i1, i2, i3, i4 and i5 are the weights of different types of eigenvalues, and i1-i5 can be set according to needs. As an example, considering that the maximum daily precipitation is the focus of flood disasters during a heavy rainfall process, the weight of the maximum daily precipitation can be set higher, for example, i1-i5 can be set to: 2, 1, 1, 1, 1, 1.

[0136] The historical heavy precipitation events whose matching degree meets the matching condition are taken as the target historical heavy precipitation events.

[0137] After determining the target historical heavy precipitation event, the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event is obtained.

[0138] The hourly rainfall pattern of historical heavy rainfall events is used to describe the hourly rainfall conditions in historical heavy rainfall events. The hourly rainfall pattern of historical heavy rainfall events includes the hourly rainfall ratio. The hourly rainfall ratio is the ratio of the hourly rainfall to the daily rainfall.

[0139] In a possible implementation, the hourly rainfall type of historical heavy precipitation events belongs to the hourly rainfall type library of historical heavy precipitation events. The hourly rainfall type library of historical heavy precipitation events is established based on each historical heavy precipitation event included in the historical heavy precipitation event library. The hourly rainfall type library of historical heavy precipitation events includes the hourly rainfall types of historical heavy precipitation events of each historical heavy precipitation event.

[0140] The hourly rainfall type of the historical heavy rainfall event of the target historical heavy rainfall event is obtained, and the precipitation data of the simulated heavy rainfall event is downscaled according to the hourly rainfall type of the historical heavy rainfall event of the target historical heavy rainfall event. The idea of ​​downscaling is to determine the hourly precipitation ratio of the precipitation data of the simulated heavy rainfall event according to the hourly rainfall type of the historical heavy rainfall event of the target historical heavy rainfall event, and then determine the hourly precipitation of the simulated heavy rainfall event according to the daily precipitation of the simulated heavy rainfall event, and obtain the hourly precipitation data.

[0141] First, simulated precipitation events are divided into simulated heavy precipitation events and simulated non-heavy precipitation events. For simulated heavy precipitation events, the hourly rainfall type of historical heavy precipitation events of the matched target historical heavy precipitation events is used to divide the precipitation data. In this way, the hourly rainfall type of historical heavy precipitation events of the target historical heavy precipitation events with similar characteristics can be used to downscale the precipitation data. On the one hand, different simulated heavy precipitation events may match different target historical heavy precipitation events, and the hourly rainfall types used for downscaling are more abundant, which enhances the randomness and flexibility of the downscaling process. On the other hand, the hourly rainfall type of historical heavy precipitation events of the matched target historical heavy precipitation events is used for downscaling, and the hourly precipitation data obtained is more effective, which is conducive to a more accurate risk simulation assessment of flood disasters.

[0142] As an example, the present application embodiment provides a possible implementation method for dividing the precipitation data of the simulated heavy precipitation event into hourly scale precipitation data according to the hourly rainfall type of the historical heavy precipitation event of the target historical heavy precipitation event. Figure 2 As shown, this figure is a schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application, wherein S103 includes S1031-S1034.

[0143] S1031: If the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event, then the precipitation days included in the simulated heavy precipitation event are matched one by one with the precipitation days included in the target historical heavy precipitation event in date order.

[0144] Determine whether the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event. If they are equal, the precipitation days included in the simulated heavy precipitation event and the precipitation days included in the target historical heavy precipitation event can be matched one by one according to the date order included in the precipitation event. For example, if the simulated heavy precipitation event is 3 days and the target historical heavy precipitation event is 3 days, the number of days of the simulated heavy precipitation event and the number of days of the target historical heavy precipitation event are equal. Match the first day of the simulated heavy precipitation event with the first day of the target historical heavy precipitation event. Match the second day of the simulated heavy precipitation event with the second day of the target historical heavy precipitation event. Match the third day of the simulated heavy precipitation event with the third day of the target historical heavy precipitation event.

[0145] S1032: According to the hourly rainfall patterns of historical heavy rainfall events, including the hourly rainfall patterns of each rainfall day, the precipitation data of the simulated heavy rainfall events matching the rainfall days are divided to obtain the hourly precipitation data of each rainfall day of the simulated heavy rainfall events.

[0146] The hourly rainfall patterns of historical heavy precipitation events include the hourly rainfall patterns of each precipitation day, and the precipitation data of the precipitation days of the matching simulated precipitation events are divided.

[0147] Take the above-mentioned simulated heavy precipitation event and the target historical heavy precipitation event as an example, both of which last for 3 days. According to the hourly rain type of the historical heavy precipitation event on the first day of the target historical heavy precipitation event, the precipitation data on the first day of the simulated precipitation event is downscaled. For example, the hourly rain type of the historical heavy precipitation event on the first day of the target historical heavy precipitation event is {a%, b%, c%, ..., x%}. The precipitation on the first day of the simulated precipitation event is L. The precipitation data on the first day of the simulated precipitation event is downscaled to obtain the precipitation amount per hour as {a% × L, b% × L, c% × L, ..., x% × L}. Similarly, according to the hourly rain type of the historical heavy precipitation event on the second day of the target historical heavy precipitation event, the precipitation data on the second day of the simulated precipitation event is downscaled. According to the hourly rain type of the historical heavy precipitation event on the third day of the target historical heavy precipitation event, the precipitation data on the third day of the simulated precipitation event is downscaled. In this way, the hourly precipitation data of the simulated precipitation event is obtained.

[0148] S1033: If the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, then determine the matching relationship between the day with the maximum precipitation in the simulated heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event, and determine the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event based on the relative order of dates. The relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation.

[0149] If the number of days of simulated heavy precipitation events is not equal to the number of days of target historical heavy precipitation events, for example, the number of days of simulated heavy precipitation events is greater than the number of days of target historical heavy precipitation events, or the number of days of simulated heavy precipitation events is less than the number of days of target historical heavy precipitation events, then the matching method based on the day with the maximum precipitation is adopted.

[0150] First, determine the matching relationship between the day with the maximum precipitation in the simulated heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event.

[0151] As an example, the simulated heavy precipitation event is 4 days, and the target historical heavy precipitation event is 3 days. The day with the maximum precipitation in the simulated heavy precipitation event is the 3rd day, and the day with the maximum precipitation in the target historical heavy precipitation event is the 2nd day, so a matching relationship is established between the 3rd day of the simulated heavy precipitation event and the 2nd day of the target historical heavy precipitation event.

[0152] Taking the day with the maximum precipitation as the benchmark, match other precipitation days in the simulated heavy precipitation event. First, determine the relative order of dates. The relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation. For example, for the simulated heavy precipitation event with the maximum precipitation day being the third day, the first day of the simulated heavy precipitation event is two days before the maximum precipitation day, the second day of the simulated heavy precipitation event is the day before the maximum precipitation day, and the fourth day of the simulated heavy precipitation event is the day after the maximum precipitation day. For the target historical heavy precipitation event with the maximum precipitation day being the second day, the first day of the target historical heavy precipitation event is the day before the maximum precipitation day, and the third day of the target historical heavy precipitation event is the day after the maximum precipitation day.

[0153] According to the relative order of dates, determine the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event. Match the precipitation days of the simulated heavy precipitation event with the precipitation days of the target historical heavy precipitation event with the same relative order of dates.

[0154] Still taking the above example, the second day of the simulated heavy precipitation event whose relative date sequence is the previous day is matched with the first day of the target historical heavy precipitation event. The fourth day of the simulated heavy precipitation event whose relative date sequence is the next day is matched with the third day of the target historical heavy precipitation event.

[0155] S1034: According to the hourly rainfall patterns of the precipitation days included in the hourly rainfall patterns of the historical heavy rainfall events, the precipitation data of the simulated heavy rainfall events matching the precipitation days are divided to obtain the hourly-scale precipitation data of the simulated heavy rainfall events.

[0156] The hourly rainfall patterns of historical heavy precipitation events include the hourly rainfall patterns of each precipitation day, and the precipitation data of the precipitation days of the matching simulated precipitation events are divided.

[0157] Take the above-mentioned simulated heavy precipitation event of 4 days and the target historical heavy precipitation event of 3 days as an example. According to the hourly rain type of the historical heavy precipitation event on the first day of the target historical heavy precipitation event, the precipitation data of the second day of the simulated precipitation event is downscaled. According to the hourly rain type of the historical heavy precipitation event on the second day of the target historical heavy precipitation event, the precipitation data of the third day of the simulated precipitation event is downscaled. According to the hourly rain type of the historical heavy precipitation event on the third day of the target historical heavy precipitation event, the precipitation data of the fourth day of the simulated precipitation event is downscaled. In this way, the hourly scale precipitation data of the simulated precipitation event is obtained.

[0158] In the case where the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, in some possible implementations, all precipitation days included in the simulated heavy precipitation event can be matched with the precipitation days of the target historical heavy precipitation event. In other possible implementations, the precipitation days included in the simulated heavy precipitation event cannot all be matched with the precipitation days of the target historical heavy precipitation event. For example, the above example of 4 days of simulated heavy precipitation events and 3 days of target historical heavy precipitation events. Day 1 of the simulated heavy precipitation event does not match a precipitation day of the target historical heavy precipitation event.

[0159] If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, the precipitation data of the unmatched precipitation days need to be downscaled.

[0160] For further information, see Figure 3 As shown, this figure is a schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application, which includes S105 in addition to the above S101-S104.

[0161] S105: If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and based on the second day sequence of the unmatched precipitation days, determine the target random hourly rainfall type corresponding to the second day sequence, and divide the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event into hourly scale precipitation data according to the target random hourly rainfall type corresponding to the second day sequence.

[0162] The second day sequence of the unmatched precipitation days is the order of the number of days of the unmatched precipitation days in a year. Get the target random hourly rain type corresponding to the second day sequence. The target random hourly rain type of the second day sequence is determined according to the precipitation data of the historical precipitation events that occurred in the second day sequence. The target random hourly rain type includes the hourly precipitation ratio.

[0163] The following is an introduction to random hourly rainfall patterns. Random hourly rainfall patterns belong to the daily sequence random hourly rainfall pattern library. The daily sequence random hourly rainfall pattern library is constructed using the following steps B1-B3:

[0164] B1: Obtain the historical precipitation data of the target site. The historical precipitation data includes the hourly rainfall type on the precipitation day.

[0165] The hourly rainfall pattern of a precipitation day includes the hourly precipitation ratio.

[0166] B2: From the historical precipitation data, obtain the hourly rainfall pattern of the precipitation days within the daily sequence range of each day sequence, and obtain the random hourly rainfall pattern of the day sequence.

[0167] The date range is a range including at least one precipitation day determined based on the date. The date range can be set according to the need to establish a random hourly rainfall type library. As an example, the date range is determined based on the date and the date distance value. For example, the date distance value is 15 days. The date range is from the first 15 days of the date to the last 15 days of the date. For example, the date is date, and the date range is [date-15, date+15].

[0168] For each date sequence, the hourly rainfall pattern of each precipitation day within the date sequence range is obtained from the historical precipitation data, and the set of hourly rainfall patterns of precipitation days within the date sequence range of each date sequence is used to form the random hourly rainfall pattern of the date sequence. If the date sequence range does not include precipitation days, the random hourly rainfall pattern of the date sequence is empty, and it can be considered that the random hourly rainfall pattern of the date sequence does not exist.

[0169] Furthermore, in order to improve the accuracy of the random hourly rainfall pattern, for each day sequence, the hourly rainfall pattern of each precipitation day whose precipitation within the day sequence range is greater than or equal to the minimum threshold is obtained. The minimum threshold is, for example, 10 mm. In this way, it can be ensured that the precipitation day has a certain precipitation amount, and the adverse effect of the precipitation data of the precipitation day with too low precipitation on the random hourly rainfall pattern can be avoided.

[0170] B3: Use the random hourly rainfall patterns of each day sequence to build a random hourly rainfall pattern library for each day sequence.

[0171] The constructed daily random hourly rainfall pattern database can be used to query the random hourly rainfall pattern corresponding to a specific daily sequence.

[0172] It should be noted that the above B1-B3 is a method for generating a daily sequence random hourly rainfall pattern library for a target site. In addition, the daily sequence random hourly rainfall pattern library may also include random hourly rainfall patterns of multiple sites. The random hourly rainfall pattern of each site can be obtained by adopting the steps of B1-B3.

[0173] First, according to the second day sequence, the random hourly rain pattern corresponding to the second day sequence is queried. The random hourly rain pattern corresponding to the second day sequence may include the random hourly rain patterns of one or more precipitation days. In the case where the random hourly rain pattern corresponding to the second day sequence includes the random hourly rain patterns of multiple precipitation days, a random hourly rain pattern of a precipitation day is randomly selected from the random hourly rain patterns of multiple precipitation days as the target random hourly rain pattern corresponding to the second day sequence. In the case where the random hourly rain pattern corresponding to the second day sequence includes the random hourly rain pattern of a precipitation day, the random hourly rain pattern of the precipitation day is used as the target random hourly rain pattern corresponding to the second day sequence.

[0174] After determining the target random hourly rainfall type corresponding to the second day sequence, the precipitation data of the unmatched precipitation days included in the simulated heavy rainfall event are divided into hourly scale precipitation data according to the target random hourly rainfall type corresponding to the second day sequence. The way of downscaling the precipitation data is similar to the way of downscaling the precipitation data in the above step B2, which will not be repeated here.

[0175] For unmatched precipitation days, the target random hourly rainfall patterns corresponding to the day sequence are used for downscaling, which can enrich the types of hourly rainfall patterns used, further improve the randomness of the generated hourly-scale precipitation patterns, and also improve the effectiveness of the generated hourly-scale precipitation patterns.

[0176] Based on the above-mentioned method for constructing a random hourly rainfall pattern library in a daily sequence, it can be known that there may be a situation where there is no target random hourly rainfall pattern corresponding to the second daily sequence.

[0177] For further information, see Figure 4 As shown, this figure is a schematic flow chart of another method for downscaling the time scale of precipitation data provided in an embodiment of the present application, which includes S106 in addition to the above S101-S105.

[0178] S106: If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and there is no target random hourly rain pattern corresponding to the second day sequence of the unmatched precipitation days, then the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event are divided into hourly scale precipitation data using the standard hourly rain pattern.

[0179] Among them, the standard hourly rainfall type is determined based on the precipitation data of historical precipitation events at the target station. The standard hourly rainfall type includes the hourly precipitation ratio.

[0180] The following is an introduction to the standard hourly rainfall pattern. The standard hourly rainfall pattern belongs to the standard hourly rainfall pattern library. The standard hourly rainfall pattern library is constructed using the following C1-C3 methods:

[0181] C1: Obtain the historical precipitation data of the target site. The historical precipitation data includes the hourly rainfall type on the precipitation day.

[0182] The hourly rainfall pattern of a precipitation day includes the hourly precipitation ratio.

[0183] C2: Based on historical precipitation data, determine the average value of the precipitation ratio of the target hour in the hourly rainfall pattern of each precipitation day as the precipitation ratio of the target hour in the standard hourly rainfall pattern.

[0184] From the historical precipitation data, the hourly rainfall pattern of each precipitation day is obtained. For the hourly rainfall pattern of each precipitation day, the average value of the precipitation ratio is determined for each hour, that is, the target hour, to obtain the precipitation ratio of the target hour in the standard hourly rainfall pattern.

[0185] As an example, the hourly rainfall pattern of each precipitation day in the historical precipitation data is obtained. For example, the precipitation ratio of the first hour in the hourly rainfall pattern of each precipitation day is averaged to obtain the average value of the precipitation ratio, which is used as the precipitation ratio of the first hour in the standard hourly rainfall pattern. By analogy, the average value of the precipitation ratio of the other 23 hours in the hourly rainfall pattern is determined as the precipitation ratio of the other 23 hours in the standard hourly rainfall pattern. In addition, the determined precipitation ratio of 24 hours is adaptively adjusted according to the ratio so that the sum of the precipitation ratios of the 24 hours is 100%.

[0186] The standard hourly rainfall pattern determined based on the precipitation data of historical precipitation events at the target station is station-dependent and has nothing to do with the day sequence.

[0187] C3: Use standard hourly rainfall patterns to build a standard hourly rainfall pattern library.

[0188] Based on the standard hourly rainfall type library constructed as above, the standard hourly rainfall type is queried. Using the standard hourly rainfall type, the precipitation data of the unmatched precipitation days included in the simulated heavy rainfall event is divided into hourly-scale precipitation data. The way of downscaling the precipitation data is similar to the way of downscaling the precipitation data in the above-mentioned step B2, which will not be repeated here. For unmatched precipitation days and target random hourly rainfall types that do not correspond to the day sequence, the standard hourly rainfall type is used for downscaling. The standard hourly rainfall type can adapt to each day sequence, realize the matching of different types of hourly rainfall types used in precipitation days in various situations, expand the types of hourly rainfall types that can be used for downscaling, improve the randomness of the generated hourly-scale precipitation rainfall type, and also improve the effectiveness of the generated hourly-scale precipitation rainfall type.

[0189] The above is a downscaling method for precipitation data of heavy precipitation events. The following is an introduction to the downscaling method for precipitation data simulating non-heavy precipitation events.

[0190] S104: If a target random hourly rainfall pattern corresponding to the first day sequence is determined based on the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event is divided into hourly-scale precipitation data according to the target random hourly rainfall pattern.

[0191] The first day sequence of the first precipitation day, the order of the days of the first precipitation day in a year. Get the target random hourly rain type corresponding to the first day sequence. The target random hourly rain type of the first day sequence is determined according to the precipitation data of the historical precipitation event that occurred in the first day sequence. The target random hourly rain type includes the hourly precipitation ratio.

[0192] Specifically, according to the first day sequence, the random hourly rain pattern corresponding to the first day sequence is queried. The random hourly rain pattern corresponding to the first day sequence may include random hourly rain patterns of one or more precipitation days. In the case where the random hourly rain pattern corresponding to the first day sequence includes random hourly rain patterns of multiple precipitation days, a random hourly rain pattern of a precipitation day is randomly selected from the random hourly rain patterns of multiple precipitation days as the target random hourly rain pattern corresponding to the first day sequence. In the case where the random hourly rain pattern corresponding to the first day sequence includes a random hourly rain pattern of a precipitation day, the random hourly rain pattern of the precipitation day is used as the target random hourly rain pattern corresponding to the first day sequence.

[0193] The target random hourly rainfall pattern of the first day sequence is determined based on the precipitation data of the historical precipitation events that occurred on the first day sequence. The target random hourly rainfall pattern includes the hourly precipitation ratio.

[0194] After determining the target random hourly rainfall type corresponding to the first day sequence, the precipitation data of the precipitation days included in the simulated non-heavy precipitation event are divided into hourly-scaled precipitation data according to the target random hourly rainfall type corresponding to the first day sequence. The way of downscaling the precipitation data is similar to the way of downscaling the precipitation data in the above step B2, which will not be repeated here.

[0195] For simulation of non-heavy precipitation events, the target random hourly rainfall patterns corresponding to the daily sequence are used for downscaling, which conforms to the characteristics of simulation of non-heavy precipitation events, improves the randomness of the generated hourly precipitation patterns, and also improves the effectiveness of the generated hourly precipitation patterns, ensuring the comprehensiveness of the precipitation event set for assessing flood disasters and improving the effect of assessing flood disasters.

[0196] Based on the above-mentioned method for constructing a random hourly rainfall pattern library in a daily sequence, it can be known that there may be a situation where the target random hourly rainfall pattern corresponding to the first daily sequence does not exist.

[0197] For further information, see Figure 4 As shown, S107 is also included.

[0198] S107: If there is no target random hourly rainfall pattern corresponding to the first day sequence, the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event is divided into hourly-scale precipitation data using the standard hourly rainfall pattern.

[0199] Among them, the standard hourly rainfall type is determined based on the precipitation data of historical precipitation events. The standard hourly rainfall type includes the hourly precipitation ratio.

[0200] Based on the above-constructed standard hourly rainfall type database, the standard hourly rainfall type is queried. Using the standard hourly rainfall type, the precipitation data of the first precipitation day included in the simulated non-heavy rainfall event is divided into hourly scale precipitation data. The way of downscaling the precipitation data is similar to the way of downscaling the precipitation data in the above-mentioned step B2, which will not be repeated here.

[0201] For precipitation days that do not have a target random hourly rainfall pattern corresponding to the daily sequence, standard hourly rainfall patterns are used for downscaling to achieve the matching of different types of hourly rainfall patterns used in precipitation days under various circumstances, expand the types of hourly rainfall patterns that can be used for downscaling, and improve the randomness of the generated hourly-scale precipitation patterns, which is conducive to maintaining the overall randomness of the precipitation event set in the long time series and improving the effectiveness of the precipitation event set in assessing flood disasters.

[0202] Furthermore, after obtaining the downscaled precipitation data of the target station, the spatial correlation of the downscaled precipitation data of multiple stations in the same area can be reconstructed.

[0203] The present application embodiment provides a method for downscaling the time scale of precipitation data, see Figure 5 As shown, in addition to steps S101-S107, the following steps are also included:

[0204] S108: Obtain downscaled precipitation data of multiple stations belonging to the same area included in the precipitation random event set, where the downscaled precipitation data includes hourly-scale precipitation data of multiple simulated precipitation events.

[0205] After downscaling the precipitation data to be processed at multiple stations in the same area in terms of time scale to obtain downscaled precipitation data, the downscaled precipitation data of multiple stations in the same area are obtained and further spatial correlation processing is performed. The downscaled precipitation data includes precipitation data at the hourly scale.

[0206] S109: Obtain precipitation data of precipitation days in the same date sequence from the downscaled precipitation data of multiple stations, and obtain precipitation data of multiple stations to be reconstructed.

[0207] The downscaled precipitation data includes multiple simulated precipitation events. From the downscaled precipitation data of multiple stations, the precipitation data of the precipitation day in the same sequence is obtained. In this way, the spatial correlation of the precipitation data of the same precipitation day can be reconstructed. The precipitation data of the precipitation day in the same sequence of multiple stations are used as the precipitation data of the station to be reconstructed.

[0208] S110: The order of the hourly precipitation ratios of the precipitation data of the to-be-reconstructed station with the largest daily precipitation is used as a standard order.

[0209] Among the precipitation data of multiple sites to be reconstructed, the precipitation data of the site to be reconstructed with the largest daily precipitation is determined. The order of the hourly precipitation ratio of the precipitation data of the site to be reconstructed with the largest daily precipitation is obtained. The order of the hourly precipitation ratio of the precipitation data of the site to be reconstructed with the largest daily precipitation is used as the standard order.

[0210] As an example, the same area includes three stations, station X, station Y, and station Z. The precipitation data of the 4th day at station X includes x(1)-x(24), the precipitation data of the 4th day at station Y includes y(1)-y(24), and the precipitation data of the 4th day at station Z includes z(1)-z(24). x(a) represents the precipitation ratio of the ath hour of the 4th day at station X. y(b) represents the precipitation ratio of the bth hour of the 4th day at station Y. z(c) represents the precipitation ratio of the cth hour of the 4th day at station Z.

[0211] Among them, the daily precipitation at station Z is the largest. The order of the hourly precipitation ratios at station Z is, for example, z(1)=z(2)=…=z(21)<z(19)<z(20). That is to say, z(20) is the largest precipitation ratio, z(19) is the second largest precipitation ratio, and the precipitation ratios of the remaining hours are equal. Equal precipitation ratios can be sorted in chronological order. The order of the hourly precipitation ratios at station Z is determined as the standard order. The standard order, that is, the order of precipitation ratios from large to small is: precipitation ratio of the 20th hour, precipitation ratio of the 19th hour, precipitation ratio of the 1st hour, precipitation ratio of the 2nd hour, ..., precipitation ratio of the 18th hour, precipitation ratio of the 21st hour, ..., precipitation ratio of the 24th hour.

[0212] S111: adjusting the order of hourly precipitation ratios in precipitation data of other stations to be reconstructed according to the standard order, to obtain reconstructed precipitation data of the station.

[0213] Using the standard order, adjust the order of the hourly precipitation ratios in the precipitation data of other stations to be reconstructed. This can make the change trend of precipitation ratios at different stations in the same area close, which is consistent with the correlation of precipitation in the same area in terms of spatial extent.

[0214] Take the above-mentioned X site, Y site and Z site as an example.

[0215] The order of the hourly precipitation ratios at station X is, for example: x(1)=x(2)=…=x(19)<x(8)<x(9). That is, x(9) is the largest precipitation ratio, x(8) is the second largest precipitation ratio, and the precipitation ratios of the remaining hours are equal. According to the standard order, the order of the hourly precipitation ratios at station X is adjusted. The largest precipitation ratio at station X (i.e., x(9)) is used as the precipitation ratio for the 20th hour, the second largest precipitation ratio at station X (i.e., x(8)) is used as the precipitation ratio for the 19th hour, and so on. The order of the precipitation ratios of the remaining hours at station X is adjusted in sequence according to the standard order. The order of the hourly precipitation ratios at station X after adjustment is the same as the standard order, and the order of precipitation ratios from large to small is: precipitation ratio in the 20th hour, precipitation ratio in the 19th hour, precipitation ratio in the 1st hour, precipitation ratio in the 2nd hour, ..., precipitation ratio in the 18th hour, precipitation ratio in the 21st hour, ..., precipitation ratio in the 24th hour.

[0216] The order of the hourly precipitation ratio of station Y is, for example: y(1)=y(2)=…=y(15)<y(16)<y(18)<y(17). That is, y(17) is the largest precipitation ratio, y(18) is the second largest precipitation ratio, y(16) is the third largest precipitation ratio, and the precipitation ratios of the remaining hours are equal. According to the standard order, the order of the hourly precipitation ratio of station Y is adjusted. The largest precipitation ratio in station Y (that is, y(17)) is used as the precipitation ratio of the 20th hour, the second largest precipitation ratio in station Y (that is, y(18)) is used as the precipitation ratio of the 19th hour, and the third largest precipitation ratio in station Y (that is, y(16)) is used as the precipitation ratio of the 1st hour. By analogy, the order of the precipitation ratios of the remaining hours in station Y is adjusted in sequence according to the standard order. The order of the hourly precipitation ratios at the adjusted Y station is the same as the standard order. The order of precipitation ratios from large to small is: precipitation ratio in the 20th hour, precipitation ratio in the 19th hour, precipitation ratio in the 1st hour, precipitation ratio in the 2nd hour, ..., precipitation ratio in the 18th hour, precipitation ratio in the 21st hour, ..., precipitation ratio in the 24th hour.

[0217] In this way, the reconstructed site precipitation data of site X, site Y and site Z are obtained.

[0218] By sequentially reconstructing the hourly precipitation data of stations in the same area, the changing trends of the hourly precipitation at stations in the same area are made close, the spatial correlation of precipitation data at various stations in the same area is enhanced, and the data quality of precipitation data included in the precipitation random event set is improved, which is conducive to a more effective assessment of flood disasters using the downscaled precipitation random event set.

[0219] Based on the method for downscaling the time scale of precipitation data provided by the above method embodiment, the embodiment of the present application also provides a device for downscaling the time scale of precipitation data. The device for downscaling the time scale of precipitation data will be described below in conjunction with the accompanying drawings.

[0220] See also Figure 6 As shown in FIG. 1 , this figure is a schematic diagram of a time scale downscaling device for precipitation data provided in an embodiment of the present application. Figure 6 As shown, the downscaling device for the time scale of the precipitation data includes:

[0221] An acquisition unit 601 is used to acquire precipitation data to be processed of a target station included in the precipitation random event set to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events;

[0222] A type determination unit 602 is used to determine, in the precipitation data to be processed, a simulated precipitation event that meets the heavy precipitation event condition as a simulated heavy precipitation event, and to determine a simulated precipitation event other than the simulated heavy precipitation event as a simulated non-heavy precipitation event;

[0223] The first downscaling unit 603 is used to determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event meets a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly-scaled precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event, wherein the historical heavy precipitation event hourly rainfall type includes an hourly precipitation ratio, and the hourly precipitation ratio is a ratio of the precipitation in each hour to the precipitation on that day;

[0224] The second downscaling unit 604 is used to determine the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, and divide the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event into hourly precipitation data according to the target random hourly rainfall type; the target random hourly rainfall type of the first day sequence is determined according to the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the target random hourly rainfall type includes the hourly precipitation ratio.

[0225] In a possible implementation, the first downscaling unit 603 is configured to determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, including:

[0226] The first downscaling unit 603 is used to extract a first eigenvalue of the simulated heavy precipitation event from the precipitation data of the simulated heavy precipitation event; determine a matching degree between the simulated heavy precipitation event and the historical heavy precipitation event according to the first eigenvalue of the simulated heavy precipitation event and the second eigenvalue of the historical heavy precipitation event, wherein the historical heavy precipitation event is a historical precipitation event of the target station that meets the heavy precipitation event condition; and use the historical heavy precipitation event whose matching degree meets the matching condition as the target historical heavy precipitation event.

[0227] In a possible implementation, the historical heavy rainfall event belongs to a historical heavy rainfall event database, and the historical heavy rainfall event database is constructed in the following manner:

[0228] Acquire historical precipitation data of a target site, wherein the historical precipitation data includes precipitation data of a plurality of historical precipitation events;

[0229] In the historical precipitation data, historical precipitation events that meet the heavy precipitation event conditions are determined as historical heavy precipitation events;

[0230] Extracting a second characteristic value of the historical heavy rainfall event;

[0231] The historical heavy rainfall event database is constructed using the second characteristic value of the historical heavy rainfall event.

[0232] In a possible implementation, the first downscaling unit 603 is used to divide the precipitation data of the simulated heavy precipitation event into hourly-scaled precipitation data according to the hourly rainfall type of the historical heavy precipitation event of the target historical heavy precipitation event, including:

[0233] The first downscaling unit 603 is used to match the precipitation days included in the simulated heavy precipitation event with the precipitation days included in the target historical heavy precipitation event in date order if the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event; divide the precipitation data of the simulated heavy precipitation event matched by the precipitation day according to the hourly rainfall type of each precipitation day including the hourly rainfall type of the historical heavy precipitation event, and obtain the hourly-scaled precipitation data of each precipitation day of the simulated heavy precipitation event; if the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, determine the The matching relationship between the day with the maximum precipitation in the simulated heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event is determined, and according to the relative order of dates, the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event is determined, and the relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation; according to the hourly rainfall type of the precipitation day included in the hourly rainfall type of the historical heavy precipitation event, the precipitation data of the simulated heavy precipitation event matched by the precipitation day is divided to obtain the hourly scale precipitation data of the simulated heavy precipitation event.

[0234] In a possible implementation manner, the device further includes:

[0235] The third downscaling unit 605 is used for, if the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and according to the second day sequence of the unmatched precipitation days, determining the target random hourly rainfall type corresponding to the second day sequence, dividing the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event into hourly-scaled precipitation data according to the target random hourly rainfall type corresponding to the second day sequence.

[0236] In a possible implementation manner, the device further includes:

[0237] The fourth downscaling unit 606 is used to divide the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event into hourly-scaled precipitation data using a standard hourly rain type if the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and there is no target random hourly rain type corresponding to the second day sequence of the unmatched precipitation days. The standard hourly rain type is determined based on the precipitation data of the historical precipitation events of the target station that occurred in the second day sequence, and the standard hourly rain type includes an hourly precipitation ratio.

[0238] In a possible implementation, the second downscaling unit 604 is configured to determine, according to a first day sequence of a first precipitation day included in the simulated non-heavy precipitation event, a target random hourly rainfall type corresponding to the first day sequence, including:

[0239] The second downscaling unit 604 is used to query the random hourly rain type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, wherein the random hourly rain type corresponding to the first day sequence includes the random hourly rain type of at least one precipitation day within the day sequence range; if the random hourly rain type corresponding to the first day sequence is queried, the random hourly rain type of any precipitation day is selected from the random hourly rain types corresponding to the first day sequence as the target random hourly rain type.

[0240] In a possible implementation, the random hourly rainfall pattern belongs to a daily sequential random hourly rainfall pattern library, and the daily sequential random hourly rainfall pattern library is constructed in the following manner:

[0241] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0242] From the historical precipitation data, obtain the hourly rainfall pattern of the precipitation day within the range of each day sequence, and obtain the random hourly rainfall pattern of the day sequence;

[0243] The random hourly rainfall pattern library of each day is constructed by using the random hourly rainfall pattern of each day.

[0244] In a possible implementation manner, the device further includes:

[0245] The fifth downscaling unit 607 is used to divide the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event into hourly-scaled precipitation data using a standard hourly rain type if there is no target random hourly rain type corresponding to the first day sequence. The standard hourly rain type is determined based on the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the standard hourly rain type includes an hourly precipitation ratio.

[0246] In a possible implementation, the standard hourly rainfall type belongs to a standard hourly rainfall type library, and the standard hourly rainfall type library is constructed in the following manner:

[0247] Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days;

[0248] According to the historical precipitation data, determine the average value of the precipitation ratio of the target hour in the hourly rainfall pattern of each precipitation day as the precipitation ratio of the target hour in the standard hourly rainfall pattern, where the target hour is each hour of the day;

[0249] The standard hourly rainfall pattern library is constructed using the standard hourly rainfall pattern.

[0250] In a possible implementation manner, the device further includes:

[0251] The spatial reconstruction unit 608 is used to obtain downscaled precipitation data of multiple stations belonging to the same area included in the precipitation random event set, wherein the downscaled precipitation data includes hourly-scale precipitation data of multiple simulated precipitation events; obtain precipitation data of precipitation days in the same sequence from the downscaled precipitation data of the multiple stations to obtain precipitation data of multiple stations to be reconstructed; take the order of the hourly precipitation ratio of the precipitation data of the station to be reconstructed with the largest daily precipitation as the standard order; adjust the order of the hourly precipitation ratio in the precipitation data of other stations to be reconstructed according to the standard order to obtain the reconstructed station precipitation data.

[0252] Based on the method for downscaling the time scale of precipitation data provided by the above method embodiment, the present application provides a device for downscaling the time scale of precipitation data, including: a processor, a memory, and a system bus;

[0253] The processor and the memory are connected via the system bus;

[0254] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes the method for downscaling the time scale of precipitation data described in any of the above embodiments.

[0255] Based on a method for downscaling the time scale of precipitation data provided in the above method embodiment, the present application provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes the method for downscaling the time scale of precipitation data described in any of the above embodiments.

[0256] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0257] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0258] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0259] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0260] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for downscaling the time scale of precipitation data, characterized in that: The method comprises: Acquire precipitation data to be processed of a target site included in the precipitation random event set to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events; In the precipitation data to be processed, the simulated precipitation events that meet the heavy precipitation event conditions are determined as simulated heavy precipitation events, and the simulated precipitation events other than the simulated heavy precipitation events are determined as simulated non-heavy precipitation events; Determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly-scale precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event, wherein the historical heavy precipitation event hourly rainfall type includes an hourly precipitation ratio, and the hourly precipitation ratio is a ratio of each hour's precipitation to the precipitation on that day; If the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy rainfall event is determined, the precipitation data of the first precipitation day included in the simulated non-heavy rainfall event is divided into hourly precipitation data according to the target random hourly rainfall type; the target random hourly rainfall type of the first day sequence is determined according to the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the target random hourly rainfall type includes the hourly precipitation ratio.

2. The method according to claim 1, characterized in that The determining of a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition comprises: Extracting a first characteristic value of the simulated heavy rainfall event from the precipitation data of the simulated heavy rainfall event; Determining a matching degree between the simulated heavy precipitation event and the historical heavy precipitation event according to the first characteristic value of the simulated heavy precipitation event and the second characteristic value of the historical heavy precipitation event, wherein the historical heavy precipitation event is a historical precipitation event of the target station that meets the heavy precipitation event condition; The historical heavy precipitation events whose matching degree meets the matching condition are taken as the target historical heavy precipitation events.

3. The method according to claim 2, characterized in that The historical heavy rainfall events belong to the historical heavy rainfall event database, and the historical heavy rainfall event database is constructed in the following manner: Acquire historical precipitation data of a target site, wherein the historical precipitation data includes precipitation data of a plurality of historical precipitation events; In the historical precipitation data, historical precipitation events that meet the heavy precipitation event conditions are determined as historical heavy precipitation events; Extracting a second characteristic value of the historical heavy rainfall event; The historical heavy rainfall event database is constructed using the second characteristic value of the historical heavy rainfall event.

4. The method according to claim 1, characterized in that: According to the hourly rainfall type of the historical heavy rainfall event of the target historical heavy rainfall event, the precipitation data of the simulated heavy rainfall event is divided into hourly precipitation data, including: If the number of days of the simulated heavy precipitation event is equal to the number of days of the target historical heavy precipitation event, then the precipitation days included in the simulated heavy precipitation event are matched one by one with the precipitation days included in the target historical heavy precipitation event in date order; According to the hourly rainfall type of historical heavy rainfall events, including the hourly rainfall type of each rainfall day, the precipitation data of the simulated heavy rainfall event matched with the rainfall day is divided to obtain the hourly precipitation data of each rainfall day of the simulated heavy rainfall event; If the number of days of the simulated heavy precipitation event is not equal to the number of days of the target historical heavy precipitation event, then determine the matching relationship between the day with the maximum precipitation in the simulated heavy precipitation event and the day with the maximum precipitation in the target historical heavy precipitation event, and determine the matching relationship between other precipitation days included in the simulated heavy precipitation event and other precipitation days included in the target historical heavy precipitation event according to the relative order of dates, wherein the relative order of dates is the relative order between each precipitation day included in the same precipitation event and the day with the maximum precipitation; According to the hourly rainfall type of the precipitation day included in the hourly rainfall type of the historical heavy rainfall event, the precipitation data of the simulated heavy rainfall event matched with the precipitation day is divided to obtain the hourly-scale precipitation data of the simulated heavy rainfall event.

5. The method according to claim 4, characterized in that The method further comprises: If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and based on the second day sequence of the unmatched precipitation days, the target random hourly rain type corresponding to the second day sequence is determined, the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event are divided into hourly scale precipitation data according to the target random hourly rain type corresponding to the second day sequence.

6. The method according to claim 4, characterized in that The method further comprises: If the precipitation days included in the simulated heavy precipitation event include unmatched precipitation days, and there is no target random hourly rain type corresponding to the second day sequence of the unmatched precipitation days, then the precipitation data of the unmatched precipitation days included in the simulated heavy precipitation event are divided into hourly-scale precipitation data using a standard hourly rain type, and the standard hourly rain type is determined based on the precipitation data of historical precipitation events at the target site, and the standard hourly rain type includes an hourly precipitation ratio.

7. The method according to claim 1, characterized in that The step of determining the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event comprises: According to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, query the random hourly rain type corresponding to the first day sequence, wherein the random hourly rain type corresponding to the first day sequence includes the random hourly rain type of at least one precipitation day within the day sequence range; If the random hourly rain pattern corresponding to the first date sequence is found, the random hourly rain pattern of any precipitation day is selected from the random hourly rain patterns corresponding to the first date sequence as the target random hourly rain pattern.

8. The method according to claim 7, characterized in that The random hourly rainfall pattern belongs to a daily sequence random hourly rainfall pattern library, and the daily sequence random hourly rainfall pattern library is constructed in the following manner: Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days; From the historical precipitation data, obtain the hourly rainfall pattern of the precipitation day within the range of each day sequence, and obtain the random hourly rainfall pattern of the day sequence; The random hourly rainfall pattern library of each day is constructed by using the random hourly rainfall pattern of each day.

9. The method according to claim 7, characterized in that: The method further comprises: If there is no target random hourly rainfall pattern corresponding to the first day sequence, the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event is divided into hourly precipitation data using a standard hourly rainfall pattern, wherein the standard hourly rainfall pattern is determined based on the precipitation data of historical precipitation events at the target station, and the standard hourly rainfall pattern includes hourly precipitation ratios.

10. The method according to claim 7, characterized in that The standard hourly rainfall type belongs to the standard hourly rainfall type library, and the standard hourly rainfall type library is constructed in the following manner: Acquire historical precipitation data of the target site, wherein the historical precipitation data includes hourly rainfall patterns on precipitation days; According to the historical precipitation data, determine the average value of the precipitation ratio of the target hour in the hourly rainfall pattern of each precipitation day as the precipitation ratio of the target hour in the standard hourly rainfall pattern, where the target hour is each hour of the day; The standard hourly rainfall pattern library is constructed using the standard hourly rainfall pattern.

11. The method according to any one of claims 1 to 10, characterized in that: The method further comprises: Acquire downscaled precipitation data of a plurality of stations belonging to the same region included in the precipitation random event set, wherein the downscaled precipitation data includes hourly-scale precipitation data of a plurality of simulated precipitation events; From the downscaled precipitation data of the multiple stations, obtaining precipitation data of precipitation days in the same date sequence, and obtaining precipitation data of multiple stations to be reconstructed; The order of the hourly precipitation ratios of the precipitation data of the station to be reconstructed with the largest daily precipitation is taken as the standard order; The order of hourly precipitation in precipitation data of other stations to be reconstructed is adjusted according to the standard order to obtain reconstructed precipitation data of the station.

12. A device for downscaling the time scale of precipitation data, characterized in that: The device comprises: An acquisition unit, used for acquiring precipitation data to be processed of a target station included in the precipitation random event set to be processed, wherein the precipitation data to be processed includes daily-scale precipitation data of a plurality of simulated precipitation events; a type determination unit, configured to determine, in the precipitation data to be processed, a simulated precipitation event that meets the heavy precipitation event condition as a simulated heavy precipitation event, and to determine a simulated precipitation event other than the simulated heavy precipitation event as a simulated non-heavy precipitation event; A first downscaling unit is used to determine a target historical heavy precipitation event of a target station whose matching degree with the simulated heavy precipitation event satisfies a matching degree condition, and divide the precipitation data of the simulated heavy precipitation event into hourly-scaled precipitation data according to the historical heavy precipitation event hourly rainfall type of the target historical heavy precipitation event, wherein the historical heavy precipitation event hourly rainfall type includes an hourly precipitation ratio, and the hourly precipitation ratio is a ratio of each hour's precipitation to the precipitation on that day; The second downscaling unit is used to determine the target random hourly rainfall type corresponding to the first day sequence of the first precipitation day included in the simulated non-heavy precipitation event, and divide the precipitation data of the first precipitation day included in the simulated non-heavy precipitation event into hourly-scaled precipitation data according to the target random hourly rainfall type; the target random hourly rainfall type of the first day sequence is determined according to the precipitation data of the historical precipitation event of the target station occurring in the first day sequence, and the target random hourly rainfall type includes the hourly precipitation ratio.

13. A device, characterized in that: include: Processor, memory, system bus; The processor and the memory are connected via the system bus; The memory is used to store one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the method according to any one of claims 1 to 11.

Images