Water vapor inversion method and system based on microwave communication link and regional climate characteristics

Abstract

The invention discloses a water vapor inversion method based on microwave communication links and regional climate characteristics. The method comprises the following steps: constructing at least one microwave communication link in a target region; in a rainfall process, monitoring attenuation data of all microwave communication links in real time; inverting original water vapor data based on the water vapor inversion model and all the attenuation data; and optimizing the original water vapor data by using a pre-trained water vapor optimization model corresponding to the target area to obtain optimized water vapor data. The invention provides a water vapor inversion method and system based on a microwave communication link and regional climate characteristics, and the method comprises the steps: firstly, carrying out the inversion of water vapor data, namely, single data, of each sub-region in a target region based on the attenuation condition of the microwave communication link in a rainfall process; and then the water vapor optimization model is utilized to optimize the single data, so that the single data can better reflect the real rainfall condition of the sub-region, and the accurate monitoring of the overall rainfall condition of the target region is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of rainfall monitoring, and specifically, to a water vapor inversion method and system based on a microwave communication link and regional climate characteristics. Background Art

[0002] Currently, the measurement of water vapor mainly relies on ground meteorological stations, radiosondes, and satellite remote sensing technologies. Although ground meteorological stations can provide accurate point measurement data, their spatial coverage is limited. Especially in areas with complex terrain, it is both difficult and costly to establish meteorological stations. Radiosondes can provide relatively accurate measurements, but the error is still relatively high, about between 10% and 15%. In addition, due to the limitation of the number of launches, they can only provide limited data. Even so, they cannot be globally promoted due to the limitation of release conditions. Although satellite technology can monitor a wider area, its accuracy is often insufficient in the monitoring of surface water vapor.

[0003] During the propagation of electromagnetic waves in space, they are affected by various factors such as scattering, reflection, and absorption. During the precipitation process, water droplets will have a relatively strong interference on the transmission of electromagnetic waves. Therefore, the monitoring of precipitation can also be achieved by using the transmission situation of electromagnetic waves. The higher the frequency of electromagnetic waves, the more easily they are affected by precipitation. Specifically, at a frequency of dozens of GHz, when the frequency of the microwave signal is close to the resonance frequency of water molecules, the incident microwave signal interacting with water molecules will attenuate. As the signal frequency increases, the degree of signal attenuation also shows an increasing trend. Therefore, the method of monitoring water vapor using communication microwave signals can provide data supplementation for the existing water vapor observation network. The main advantage of this technology is that the signals in these links propagate along fixed, near-ground line-of-sight paths, which enables them to be used to monitor the atmospheric conditions near the surface. Moreover, the effective monitoring of water vapor density can be achieved by using the existing widely distributed communication network without additional large-scale infrastructure investment.

[0004] However, in the prior art, there are still certain deficiencies in the method of inverting water vapor data based on microwave communication links. The main problem is that the formula model used for inversion is obtained by data fitting based on rainfall data and link attenuation, rather than being derived from precise theoretical calculations. Therefore, there is an inevitable problem of poor accuracy that cannot be eliminated. Summary of the Invention

[0005] To address the deficiencies in the prior art, the present invention provides a water vapor retrieval method and system based on microwave communication links and regional climate characteristics. First, based on the attenuation of microwave communication links during rainfall, the water vapor data of each sub-region in the target area, i.e., monomer data, is retrieved. Then, the water vapor optimization model is used to optimize the monomer data so that the monomer data can better reflect the actual rainfall situation in the sub-region, thereby achieving accurate monitoring of the overall rainfall situation in the target area.

[0006] To achieve the above object, the specific solution adopted by the present invention is as follows: A water vapor retrieval method based on microwave communication links and regional climate characteristics, comprising the following steps: Construct at least one microwave communication link in the target area; During rainfall, continuously monitor the attenuation data of all microwave communication links; Retrieve the original water vapor data based on the water vapor retrieval model and all attenuation data; Use the pre-trained water vapor optimization model corresponding to the target area to optimize the original water vapor data to obtain optimized water vapor data; The method for training the water vapor optimization model includes: Determine the regional climate characteristics of the target area and obtain the regional rainfall data of the target area over multiple time periods; Determine at least one similar area with similar regional climate characteristics to the target area and obtain the similar rainfall data of the similar area over multiple time periods; Use the similar rainfall data to expand the regional rainfall data to obtain a sample set; Construct a basic neural network model; Use the sample set to train the basic neural network model to obtain a water vapor optimization model.

[0007] As a further optimization of the above water vapor retrieval method based on microwave communication links and regional climate characteristics: The method for constructing at least one microwave communication link in the target area includes: Determine the commercial microwave systems in the target area and determine multiple alternative links based on the commercial microwave systems; Select at least one reference link from all alternative links based on the spatial distribution of the alternative links; If the spatial distribution of all reference links meets the pre-determined monitoring requirements, use the reference link as the microwave communication link; otherwise, construct at least one new link and combine the new link with the reference link as the microwave communication link.

[0008] As a further optimization of the above water vapor inversion method based on microwave communication links and regional climate characteristics: The method for inversing the original water vapor data based on the water vapor inversion model and all attenuation data includes: Inversely calculate the attenuation data one by one based on the water vapor inversion model to obtain multiple single data; Fuse the single data based on the spatial distribution data of the microwave communication link to obtain the original water vapor data.

[0009] As a further optimization of the above water vapor inversion method based on microwave communication links and regional climate characteristics: The method for inversing the attenuation data based on the water vapor inversion model is: A w = 0.182fN”(p,T,e,f)L, where A w is the attenuation data, f is the communication frequency of the microwave communication link, p is the rain area air pressure, T is the rain area temperature, and e is the water vapor density.

[0010] As a further optimization of the above water vapor inversion method based on microwave communication links and regional climate characteristics: The method for determining multiple similar regions similar to the regional climate characteristics of the target region includes: Select multiple alternative regions with the same regional climate characteristics as the target region; Conduct the first screening of the alternative regions based on the geographical range similarity between the alternative regions and the target region; Conduct the second screening of the alternative regions based on the geographical distance between the alternative regions and the target region; Conduct the third screening of the alternative regions based on whether there is a rainfall history record in the alternative regions to obtain the similar regions.

[0011] As a further optimization of the above water vapor inversion method based on microwave communication links and regional climate characteristics: The method for data expansion of the regional rainfall data using similar rainfall data includes: Check the distribution of the regional rainfall data to obtain the insufficient part; Generate filling data corresponding to the insufficient part based on the similar rainfall data; Place the filling data into the regional rainfall data to complete the data expansion of the regional rainfall data.

[0012] As a further optimization of the above water vapor inversion method based on microwave communication links and regional climate characteristics: When there are at least two similar regions, the method for generating filling data corresponding to the insufficient part based on the similar rainfall data includes: Determine the matching degree between the similar region and the target region; Weight the similar rainfall data based on the matching degree; Extract the basic data corresponding to the insufficient part from the weighted similar rainfall data; Average the basic data to obtain the filling data.

[0013] A water vapor inversion system based on a microwave communication link and regional climate characteristics, comprising: A microwave communication module for constructing at least one microwave communication link in a target area; A data acquisition module for real-time monitoring of the attenuation data of all microwave communication links, and obtaining the regional rainfall data of the target area in multiple time periods and the similar rainfall data of a similar area in multiple time periods; A data processing module for inversing the original water vapor data based on a water vapor inversion model and all the attenuation data; A model operation module for optimizing the original water vapor data by using a pre-trained water vapor optimization model corresponding to the target area to obtain optimized water vapor data. Brief Description of the Drawings

[0014] Figure 1 is a flowchart of the method of the present invention; Figure 2 is a schematic diagram of the training method of the water vapor optimization model. Detailed Description of the Invention

[0015] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0016] As Figure 1 and 2 shown, a water vapor inversion method based on a microwave communication link and regional climate characteristics includes S1 to S4.

[0017] S1. Construct at least one microwave communication link in the target area. The method of constructing at least one microwave communication link in the target area includes S11 to S13.

[0018] S11. Determine the commercial microwave system in the target area, and determine multiple alternative links based on the commercial microwave system.

[0019] S12. Select at least one reference link from all the alternative links based on the spatial distribution of the alternative links.

[0020] S13. If the spatial distribution of all reference links meets the pre-determined monitoring requirements, the reference links are taken as the microwave communication links; otherwise, at least one new link is constructed, and the new link and the reference links are combined as the microwave communication links.

[0021] In S1, selecting some existing microwave communication links from the existing commercial microwave systems can reduce the number of microwave communication links that need to be rebuilt, thereby reducing the hardware cost. Further, if there are not enough microwave communication links that can meet the requirements selected from the commercial microwave systems, some new links need to be rebuilt. Since the target area to be monitored is usually an administrative area at the county or city level, there are differences in rainfall conditions at different locations. In order to fully monitor the target area, the main basis for judging whether the selected microwave communication links can meet the requirements is the spatial distribution, that is, to judge whether the selected microwave communication links can cover the target area. If they can cover the target area, the selected alternative links can be directly used; otherwise, new links need to be constructed. The method for judging whether the microwave communication links can cover the target area is as follows: First, divide the target area into multiple sub-areas, and then determine the corresponding relationship between the microwave communication links and the sub-areas. If at least one microwave communication link can cross any one sub-area, it means that the determined microwave communication links can already cover the target area. When dividing the sub-areas, the sub-areas can be set as squares and arranged in a matrix form.

[0022] S2. During the rainfall process, the attenuation data of all microwave communication links are monitored in real time.

[0023] S3. Based on the water vapor inversion model and all the attenuation data, the original water vapor data is inverted. The method for inverting the original water vapor data based on the water vapor inversion model and all the attenuation data includes S31 to S32.

[0024] S31. Based on the water vapor inversion model, the attenuation data is inverted one by one to obtain multiple single-body data. Since each microwave communication link is independent and spans different sub-areas, it is necessary to invert the attenuation data of each microwave communication link to obtain the single-body data of several sub-areas related to the microwave communication link, so as to determine the rainfall conditions in these sub-areas.

[0025] The method for inverting the attenuation data based on the water vapor inversion model is: A w = 0.182fN”(p, T, e, f)L, where A w is the attenuation data, f is the communication frequency of the microwave communication link, p is the rain area air pressure, T is the rain area temperature, e is the water vapor density, and the rain area air pressure and rain area temperature can be monitored by existing monitoring equipment, which belongs to the conventional technology in this field.

[0026] S32. Fuse the monomer data based on the spatial distribution data of the microwave communication links to obtain the original water vapor data. Specifically, if a sub-region is simultaneously crossed by at least two microwave communication links, the monomer data derived from the attenuation data of all microwave communication links associated with this sub-region can be averaged, so that ultimately each sub-region corresponds to a monomer data. Integrating all the monomer data together can obtain the original water vapor data.

[0027] S4. Optimize the original water vapor data using the pre-trained water vapor optimization model corresponding to the target region to obtain the optimized water vapor data. Since the water vapor inversion model is obtained by data fitting based on the existing rainfall data and the attenuation data of the microwave communication links, although the original water vapor data can be inverted, there will be a difference between the original water vapor data and the real rainfall data, resulting in that the original water vapor data can only roughly reflect the rainfall situation in the target region and cannot accurately reflect the rainfall situation in the sub-regions. And because the error of data fitting cannot be eliminated, it is also impossible to improve the accuracy of the original water vapor data by optimizing the water vapor inversion model. To solve this problem, the present invention further uses the water vapor optimization model to optimize the original water vapor data, mainly optimizing the monomer data to obtain more accurate optimized water vapor data, and the optimized water vapor data can more accurately reflect the rainfall situation in the sub-regions. When the water vapor optimization model optimizes the monomer data, it may lower or raise the monomer data.

[0028] In the present invention, the water vapor optimization model adopts a neural network model. For example, a Recurrent Neural Network (RNN) model can be used. The specific structure and working principle of this model are prior art and will not be elaborated here. To ensure the accuracy of the output result of the neural network model, a large amount of actual data needs to be used to form a data set to train the neural network model. In the present invention, the data set to be used is mainly real rainfall data. However, in actual situations, it is difficult to obtain a data set of sufficient scale for a target region. The main reason is the lack of real rainfall data corresponding to the sub-regions. To solve this problem, the present invention expands the data set based on the regional climate characteristics, so that the scale of the data set can meet the training requirements of the neural network model.

[0029] Specifically, the method for training the water vapor optimization model includes T1 to T5.

[0030] T1. Determine the regional climate characteristics of the target region and obtain the regional rainfall data of the target region in multiple time periods.

[0031] T2. Determine at least one similar region with regional climate characteristics similar to those of the target region, and obtain similar rainfall data for the similar region over multiple time periods.

[0032] For any target region, it will have its own regional climate characteristics, which can be directly represented by climate types in meteorology, such as tropical rainforest climate, tropical savannah climate, and subtropical monsoon climate, etc. Further, regions with the same regional climate characteristics will have similar rainfall patterns, that is, when rainfall occurs in these regions, the actual rainfall data is similar. Therefore, the similar rainfall data of the similar region can be combined with the regional rainfall data of the target region to achieve the expansion of the regional rainfall data.

[0033] Further, the methods for determining multiple similar regions with regional climate characteristics similar to those of the target region include T21 to T24.

[0034] T21. Select multiple alternative regions with the same regional climate characteristics as the target region.

[0035] T22. Conduct the first screening of the alternative regions based on the geographical range similarity between the alternative regions and the target region.

[0036] T23. Conduct the second screening of the alternative regions based on the geographical distance between the alternative regions and the target region.

[0037] T24. Conduct the third screening of the alternative regions based on whether there is a rainfall history record in the alternative regions to obtain the similar regions.

[0038] In T2, first, determine the alternative regions based on whether the regional climate characteristics are the same. All the determined alternative regions have the same regional climate characteristics as the target region. Therefore, the actual rainfall data in the alternative regions is more likely to be similar to the actual rainfall data in the alternative regions. Then, conduct the second screening according to the geographical distance, and select the alternative regions with a geographical distance less than the preset threshold from the target region. The selected alternative regions are closer to the target region, and the actual rainfall data is also more similar. Finally, conduct the third screening based on the rainfall history record, and eliminate the alternative regions without rainfall history records or with scarce rainfall history records, and only retain the alternative regions with relatively sufficient rainfall history records as the similar regions, so as to be able to use the similar rainfall data of these similar regions to expand the regional rainfall data of the target region, ensuring that after data expansion, the reliability of the regional rainfall data is stronger.

[0039] T3. Use the similar rainfall data to perform data expansion on the regional rainfall data to obtain a sample set. The methods for using the similar rainfall data to perform data expansion on the regional rainfall data include T31 to T33.

[0040] T31. Conduct a distribution check on the regional rainfall data to obtain the insufficient part. The insufficient part corresponds to the sub-regions, that is, first determine which sub-regions' regional rainfall data are exactly missing.

[0041] T32. Generate filling data corresponding to the insufficient part based on similar rainfall data. When there are at least two similar regions, the method of generating filling data corresponding to the insufficient part based on similar rainfall data includes T321 to T324.

[0042] T321. Determine the matching degree between the similar region and the target region. The matching degree is used to reflect the similarity between the similar region and the target region. The higher the similarity between the similar region and the target region, the closer the similar rainfall data is to the real rainfall data of the target region. The matching degree can be determined based on the longitude, latitude and geographical distance between the similar region and the target region. The closer the longitude and latitude are and the closer the geographical distance is, the higher the matching degree between the similar region and the target region.

[0043] T322. Weight the similar rainfall data based on the matching degree.

[0044] T323. Extract the part corresponding to the insufficient part from the weighted similar rainfall data as the basic data.

[0045] T324. Average the basic data to obtain the filling data.

[0046] In T322 to T324, the filling data is obtained by the method of weighted average, which can avoid the increase of the error of the filling data caused by the error of a single basic data, thus improving the reliability of the filling data, and further improving the reliability of the real rainfall data after data expansion.

[0047] T33. Place the filling data into the regional rainfall data to complete the data expansion of the regional rainfall data.

[0048] T4. Construct a basic neural network model.

[0049] T5. Use the sample set to train the basic neural network model to obtain a water vapor optimization model. The sample set can be divided into a training set, a test set and a validation set according to the ratio of 8:1:1. The specific training method belongs to conventional technology and will not be elaborated here.

[0050] Based on the water vapor optimization model trained by the above method, it can correct the single data to make it closer to the real rainfall data that may occur in the sub-region, so that the optimized water vapor data can better reflect the actual rainfall situation of each sub-region in the target region.

[0051] The present invention further provides a water vapor inversion system based on a microwave communication link and regional climate characteristics, including a microwave communication module, a data acquisition module, a data processing module, and a model operation module.

[0052] The microwave communication module is used to construct at least one microwave communication link in the target area.

[0053] The data acquisition module monitors the attenuation data of all microwave communication links in real time, and acquires the regional rainfall data of the target area in multiple time periods and the similar rainfall data of the similar area in multiple time periods.

[0054] The data processing module is used to invert the original water vapor data based on the water vapor inversion model and all the attenuation data.

[0055] The model operation module is used to optimize the original water vapor data by using the pre-trained water vapor optimization model corresponding to the target area to obtain the optimized water vapor data.

[0056] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system and device can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. In several embodiments provided in the present disclosure, it should be understood that the disclosed system, device, and method can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces, and the indirect coupling or communication connection of the device or module may be in an electrical, mechanical, or other form.

[0057] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0058] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A water vapor inversion method based on microwave communication links and regional climate characteristics, characterized in that: The steps include: constructing at least one microwave communication link in the target area; During rainfall, the attenuation data of all microwave communication links are monitored in real time; Retrieve the original water vapor data based on the water vapor inversion model and all the attenuation data; The original water vapor data is optimized using a pre-trained water vapor optimization model corresponding to the target area to obtain optimized water vapor data; The method of training the water vapor optimization model comprises: Determine the regional climate characteristics of the target area and obtain regional rainfall data for the target area over multiple time periods; Determine at least one similar region having regional climate characteristics similar to those of the target region, and obtain similar rainfall data for the similar region over multiple time periods; Using similar rainfall data to expand regional rainfall data, a sample set is obtained; Build a basic neural network model; The sample set is used to train the basic neural network model to obtain the water vapor optimization model.

2. The water vapor inversion method based on microwave communication links and regional climate characteristics according to claim 1, characterized in that: The method of establishing at least one microwave communication link in a target area comprises: determining a commercial microwave system in a target area, and determining a plurality of candidate links based on the commercial microwave system; Selecting at least one reference link from all candidate links based on the spatial distribution of the candidate links; If the spatial distribution of all reference links meets the predetermined monitoring requirements, the reference links are used as the microwave communication links; otherwise, at least one new link is constructed, and the new link is combined with the reference link as a microwave communication link.

3. The water vapor inversion method based on microwave communication links and regional climate characteristics according to claim 1, characterized in that: The methods for retrieving the original water vapor data based on the water vapor inversion model and all the attenuation data include: Based on the water vapor inversion model, the attenuation data are inverted one by one to obtain multiple monomer data; The monomer data are fused based on the spatial distribution data of the microwave communication link to obtain the original water vapor data.

4. The water vapor inversion method based on microwave communication links and regional climate characteristics as claimed in claim 3, characterized in that: The method for inverting attenuation data based on the water vapor inversion model is: A w =0.182fN”(p,T,e,f)L, Among them, A w is the attenuation data, f is the communication frequency of the microwave communication link, p is the air pressure in the rain area, T is the temperature in the rain area, and e is the water vapor density.

5. The water vapor inversion method based on microwave communication links and regional climate characteristics according to claim 1, characterized in that: Methods for identifying multiple similar areas with similar regional climate characteristics to the target area include: Select multiple candidate areas with the same regional climate characteristics as the target area; Perform a first screening of candidate areas based on the similarity of their geographical scope to the target area; Conduct a second screening of candidate areas based on the geographical distance between the candidate areas and the target area; The candidate areas were screened for the third time based on whether there was a rainfall history in the candidate areas, and similar areas were obtained.

6. The water vapor inversion method based on microwave communication links and regional climate characteristics according to claim 1, characterized in that: Methods for expanding regional rainfall data using similar rainfall data include: The regional rainfall data were checked for distribution and the insufficient part was obtained; generating filling data corresponding to the insufficient part based on similar rainfall data; The filling data is placed into the regional rainfall data to complete the data expansion of the regional rainfall data.

7. The water vapor inversion method based on microwave communication links and regional climate characteristics according to claim 6, characterized in that: When there are at least two similar regions, the method for generating filling data corresponding to the insufficient portion based on similar rainfall data includes: determining a matching degree between the similar region and the target region; Weighting similar rainfall data based on the matching degree; Extract the data corresponding to the insufficient part from the weighted similar rainfall data as basic data; The basic data are averaged to obtain the filling data.

8. A water vapor inversion system based on microwave communication links and regional climate characteristics, characterized in that: include: A microwave communication module, used to establish at least one microwave communication link in the target area; The data acquisition module monitors the attenuation data of all microwave communication links in real time, and obtains the regional rainfall data of the target area in multiple time periods and the similar rainfall data of similar areas in multiple time periods; A data processing module is used to invert the original water vapor data based on the water vapor inversion model and all the attenuation data; The model operation module is used to optimize the original water vapor data using a pre-trained water vapor optimization model corresponding to the target area to obtain optimized water vapor data.

Images