Microwave radar-based water and rainfall condition integrated sensing optimization method and system

Abstract

The application discloses a microwave radar-based water and rain condition integrated sensing optimization method and system, and relates to the technical field of hydrology and meteorology. The method comprises the following steps: monitoring low signal-to-noise ratio radar echo by using a time domain cumulative algorithm, and outputting water and rain condition time series data; correcting the time series data based on ultrasonic wave data, and outputting optimized data; calling water and rain condition related data by linking multiple spatial related sites; correcting the optimized data and the related data according to topographic topological parameters, and obtaining multi-station joint defense flow prediction values; and performing disaster risk prediction according to the multi-station joint defense flow prediction values, and outputting early warning instructions. The technical problems that the water and rain condition monitoring precision is low, the data consistency is poor, and multi-station collaborative flow prediction is difficult to realize in a low signal-to-noise ratio, complex terrain and strong interference environment are solved, the technical effect that the quality and stability of water and rain condition data in a low signal-to-noise ratio environment are improved through multi-source sensing data fusion and multi-station joint defense collaborative correction, and high-precision distributed prediction of basin section flow is realized is achieved.

Description

Technical Field

[0001] This invention relates to the field of hydrology and meteorology, specifically to an integrated sensing optimization method and system for water and rainfall conditions based on microwave radar. Background Technology

[0002] With the increasing frequency of extreme weather events, sudden hydrological disasters such as rainstorms and floods have placed higher demands on urban safety, river basin flood control, and water resource allocation. Water level, flow velocity, and rainfall information are crucial foundational data for flood prevention, disaster reduction, and early warning. The real-time nature, continuity, and accuracy of this data directly impact the effectiveness of risk assessment and decision-making.

[0003] Existing water and rainfall monitoring systems mostly employ single or distributed sensing devices such as water level gauges, rain gauges, and current meters, which suffer from limited deployment density, high maintenance costs, and insufficient spatial representativeness. In recent years, microwave radar has been gradually applied to water level, flow velocity, and rainfall monitoring due to its advantages such as non-contact operation, all-weather operation, and fast response speed. However, under complex meteorological conditions, heavy rainfall, clutter interference, or low signal-to-noise ratio scenarios, radar echoes are easily affected by wind disturbance, terrain reflection, and environmental noise, resulting in large fluctuations and decreased reliability of monitoring data.

[0004] Meanwhile, traditional hydrological and rainfall analysis methods mostly rely on single-site data, lacking collaborative modeling of upstream and downstream stations and spatially related stations. This makes it difficult to reflect the true cross-sectional flow distribution and overall basin evolution, resulting in insufficient accuracy in multi-station joint prevention flow forecasting and disaster warning range determination. Furthermore, existing early warning mechanisms are typically based on a single threshold trigger, ignoring the coupling relationship between water level change rate and rainfall intensity, thus requiring improvement in the timeliness and spatial targeting of early warnings. Summary of the Invention

[0005] This application provides an integrated sensing optimization method and system for water and rainfall conditions based on microwave radar, which solves the technical problems of low accuracy and poor data consistency in water and rainfall monitoring under low signal-to-noise ratio, complex terrain and strong interference environments, making it difficult to achieve multi-site collaborative flow prediction.

[0006] The first aspect of this application provides a method for integrated water and rainfall information sensing optimization based on microwave radar, the method comprising: A time-domain cumulative algorithm is used for real-time monitoring of low signal-to-noise ratio scenarios of water and rainfall radar echo energy at target monitoring stations. Based on the monitoring results, radar signal anti-interference enhancement is performed, and water and rainfall time-series data is output. The water and rainfall time-series data is directionally corrected based on ultrasonic wind speed and direction data, outputting optimized water and rainfall data. Multiple spatially associated stations of the target monitoring station are linked to retrieve multiple water and rainfall correlation data. Based on the terrain topology parameters of the target monitoring station and multiple spatially associated stations, distributed cross-sectional flow correction is performed on the optimized water and rainfall data and multiple water and rainfall correlation data to obtain multi-station joint prevention flow prediction values. Based on the multi-station joint prevention flow prediction values, scenario-based disaster risk prediction is performed, and a disaster warning command is output, wherein the disaster warning command includes a disaster warning range identifier.

[0007] A second aspect of this application provides an integrated water and rainfall information sensing and optimization system based on microwave radar, the system comprising: The system comprises the following modules: Radar Signal Detection Module: Employs a time-domain accumulation algorithm for real-time monitoring of low signal-to-noise ratio (SNR) radar echo energy at the target monitoring station, enhances radar signal anti-interference capabilities based on monitoring results, and outputs time-series water and rainfall data; Orientation Correction Module: Performs orientation correction on the water and rainfall time-series data based on ultrasonic wind speed and direction data, outputting optimized water and rainfall data; Data Retrieval Module: Links multiple spatially associated stations of the target monitoring station to retrieve multiple water and rainfall-related data; Cross-sectional Flow Correction Module: Performs distributed cross-sectional flow correction on the optimized water and rainfall data and multiple water and rainfall-related data based on the terrain topology parameters of the target monitoring station and multiple spatially associated stations, obtaining multi-station joint prevention flow prediction values; Risk Prediction Module: Based on the multi-station joint prevention flow prediction values, performs scenario-based disaster risk prediction and outputs disaster warning commands, wherein the disaster warning commands include a disaster warning range identifier.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, a time-domain cumulative algorithm is used to monitor the radar echo energy of the target monitoring stations in real time, especially in low signal-to-noise ratio environments. Anti-interference techniques are employed to enhance the signal, generating time-series data on water and rainfall conditions. Then, ultrasonic data is used to perform targeted corrections on this time-series data to optimize accuracy. Next, multiple related monitoring stations are linked, integrating water and rainfall data from different locations. By considering terrain and topological features, cross-sectional flow corrections are applied to this data to obtain flow predictions for multi-station joint prevention and control. Finally, based on these flow predictions, a pre-disaster risk assessment is conducted, and pre-disaster warning instructions with warning ranges are generated to issue timely alerts in response to potential disasters. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic diagram of the integrated water and rainfall sensing optimization method based on microwave radar provided in this application embodiment.

[0011] Figure 2 A schematic diagram of the structure of the integrated water and rainfall sensing optimization system based on microwave radar provided in this application embodiment.

[0012] Explanation of reference numerals in the attached diagram: Radar signal detection module 11, orientation correction module 12, data retrieval module 13, cross-sectional flow correction module 14, risk prediction module 15. Detailed Implementation

[0013] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0014] Example 1, as Figure 1 As shown, this application provides an integrated water and rainfall information sensing optimization method based on microwave radar, the method including: A time-domain accumulation algorithm is used to monitor the low signal-to-noise ratio scenario of water and rainfall radar echo energy at the target monitoring station in real time. Based on the monitoring results, radar signal anti-interference enhancement is performed, and water and rainfall time series data are output.

[0015] In this embodiment, the microwave radar at the target monitoring station first continuously transmits electromagnetic waves and receives echo signals according to a preset sampling period. The received raw echo signals in each sampling period are converted from analog to digital to obtain a corresponding digital echo signal sequence. Subsequently, energy calculation is performed on the echo signal sequence, i.e., the total energy of the echo signal per unit time is calculated to obtain the radar echo energy value corresponding to each sampling period. The echo energy values ​​of multiple consecutive sampling periods are stored in a time-series buffer. In the time-domain cumulative processing, the system selects the echo energy values ​​of N consecutive sampling periods using a sliding time window, sums or performs weighted summation to obtain the cumulative echo energy value. This cumulative processing improves the identifiability of weak echoes in the time dimension. Simultaneously, the cumulative echo energy value is compared with a preset low signal-to-noise ratio (SNR) threshold. When the cumulative echo energy value is lower than the low SNR threshold, the current target monitoring station is determined to be in a low SNR water and rainfall monitoring scenario. Upon identifying a low signal-to-noise ratio (SNR) scenario, the system automatically switches to anti-interference enhancement mode. This involves extending the sampling time of the original echo signal sequence, increasing the effective time window of a single observation while maintaining the radar transmit power, to obtain more echo samples. Then, within the extended sampling time window, multiple echo signals are superimposed, and moving average filtering, low-pass filtering, and static clutter suppression are sequentially applied to the superimposed echo signal to reduce interference components from environmental noise, wind and rain scattering, and non-target reflectors. After anti-interference enhancement, the corresponding water level, flow velocity, and radar rainfall values ​​are extracted from the enhanced echo signal and continuously output in chronological order as water level time-series data, flow velocity time-series data, and radar rainfall time-series data. Finally, the water level time-series data, flow velocity time-series data, and radar rainfall time-series data are unified and output as hydrological and rainfall time-series data, providing a stable and traceable data foundation for subsequent wind disturbance correction, multi-station joint prevention and control flow prediction, and pre-disaster risk analysis.

[0016] Based on ultrasonic wind speed and direction data, the time series data of water and rainfall conditions are directionally corrected to output optimized water and rainfall conditions data.

[0017] In one embodiment, firstly, an ultrasonic anemometer is installed to collect wind speed and direction data in real time. The ultrasonic anemometer measures wind speed and direction using the principle of high-frequency sound wave propagation, acquiring real-time wind speed data (unit: m / s) and wind direction data (unit: degrees), which are then converted into a time-series format to form ultrasonic anemometer and direction data. This data includes information on wind strength and direction angle at different time periods. Subsequently, the hydrological and rainfall time-series data is matched and synchronized with the ultrasonic anemometer and direction data. The wind speed and direction data are matched one-to-one with the corresponding hydrological and rainfall time-series data according to timestamps to ensure that the wind speed and direction data belong to the same time period. Based on the synchronized data, the hydrological and rainfall time-series data is directionally corrected according to the relationship between wind direction and water flow direction, completing the wind disturbance error compensation operation, thereby forming corrected flow velocity data. Finally, the corrected flow velocity data, together with the water level time series data and radar rainfall time series data in the hydrological and rainfall time series data, will constitute the optimized hydrological and rainfall data. This optimized hydrological and rainfall data can be used for further flow forecasting, risk analysis and early warning, to ensure that hydrological forecasting and disaster prevention under complex meteorological conditions can be based on more accurate data for decision-making.

[0018] Furthermore, the method also includes: Flow velocity time-series data is extracted from the water and rainfall time-series data, which includes water level time-series data, flow velocity time-series data, and radar rainfall time-series data. Wind disturbance error compensation is performed on the flow velocity time-series data based on the ultrasonic wind speed and direction data to obtain corrected flow velocity data. The corrected flow velocity data, water level time-series data, and radar rainfall time-series data constitute optimized water and rainfall data.

[0019] Preferably, the system first extracts flow velocity time-series data from the hydrological and rainfall time-series data. This data includes water level time-series data, flow velocity time-series data, and radar rainfall time-series data. The extracted flow velocity time-series data contains the flow velocity values ​​at different time points of the target monitoring station. These flow velocity values ​​are raw data measured by flow velocity radar. Subsequently, the system compensates for wind disturbance errors in the flow velocity time-series data based on wind speed and direction data provided by ultrasonic anemometers. Specifically, it compares wind speed and direction with the flow velocity direction to determine the direction and magnitude of the wind's effect on the flow velocity. When the wind speed is in the same direction as the water flow, the wind's thrust accelerates the water flow, leading to an increase in flow velocity; conversely, if the wind speed is in the opposite direction to the water flow, the wind's effect obstructs the water flow, leading to a decrease in flow velocity. Based on the relationship between wind speed and direction, the system calculates the wind's influence on the flow velocity and applies it to the flow velocity time-series data to correct for wind disturbance at each time point. By correcting the flow velocity time-series data, the obtained corrected flow velocity data more accurately reflects the changes in water flow under actual meteorological conditions. Subsequently, the corrected flow velocity data is combined with water level time-series data and radar rainfall time-series data to form complete optimized hydrological and rainfall data. Compared to the original hydrological and rainfall data, this optimized data takes into account the interference of wind speed and direction, providing more accurate and reliable hydrological monitoring results. It can be further used for subsequent work such as multi-station joint prevention and control flow forecasting, pre-disaster risk analysis, and early warning command generation, ensuring higher monitoring precision and forecast accuracy.

[0020] Furthermore, the method also includes: The first echo energy value of the water level radar is monitored in real time. If the first echo energy value is lower than a preset first threshold, a low signal-to-noise ratio water level monitoring scenario is determined. In the low signal-to-noise ratio water level monitoring scenario, the sampling time delay of the water level radar is applied, and an extended sampling time window is set. The echo signal sequence of the water level radar is subjected to superposition and moving average filtering processing in the extended sampling time window, and the water level time series data is output.

[0021] Optionally, the system first monitors the echo signals received by the water level radar in real time. The intensity of these echoes reflects the water level height and changes within the monitoring area. The first echo energy value, i.e., the radar signal strength received during the initial sampling period, is calculated from the real-time acquired echo signals. Subsequently, the obtained first echo energy value is compared with a preset first threshold. If this first echo energy value is lower than the set first threshold, the current monitoring scenario is determined to be a low signal-to-noise ratio (SNR) water level monitoring scenario. This scenario typically occurs under environmental conditions such as rain, fog, clutter, or long-distance observation, which may reduce the effectiveness of the radar signal. Once a low SNR water level monitoring scenario is determined, the system automatically switches to an anti-interference enhancement mode. In this mode, to improve the SNR of the echo signal, the system extends the time window length for each sampling, increasing the acquisition time of the effective signal. This extended sampling process allows the radar to receive more echo signal samples, thereby increasing the ratio of the target signal to noise and increasing the stability of the radar signal. After extending the sampling time window, signal processing is performed on the acquired echo signal sequence. Specifically, the system accumulates echo signals from multiple adjacent sampling periods through superposition processing to enhance the strength of the effective echo signal. Then, a moving average filtering method is used to calculate the average value within a certain range around each sampling point, smoothing the echo signal and reducing the impact of random noise and short-term fluctuations. This approach further enhances signal quality, ensuring more stable and reliable water level data. After superposition and filtering, the output water level time-series data will be more accurate, effectively reflecting the true situation of water level changes and reducing the impact of low signal-to-noise ratio environments on the accuracy of monitoring data. Ultimately, this processed water level time-series data will serve as the data foundation for subsequent analysis and decision support applications such as flow forecasting, risk assessment, and early warning systems.

[0022] Furthermore, based on the ultrasonic wind speed and direction data, wind disturbance error compensation is performed on the flow velocity time series data to obtain corrected flow velocity data. The method further includes: Based on the installation direction of the flow velocity radar equipment at the target monitoring station, the reference angle of the water flow direction is determined; the flow velocity time series data is traversed using a preset flow velocity threshold to locate multiple flow velocity time series segments to be compensated; based on the start and end timestamps of the multiple flow velocity time series segments to be compensated, multiple wind speed and direction segments are segmented from the ultrasonic wind speed and direction data; vector compensation calculations are performed on the multiple flow velocity time series segments to be compensated using the multiple wind speed and direction segments to obtain multiple corrected flow velocity time series segments; based on the start and end timestamps of the multiple flow velocity time series segments to be compensated, the multiple corrected flow velocity time series segments are smoothly replaced in the flow velocity time series data to complete wind disturbance error compensation and obtain the corrected flow velocity data.

[0023] Optionally, the installation direction of the flow velocity radar equipment at the target monitoring station is first obtained. The installation direction of the flow velocity radar equipment determines the angular relationship between the measurement data and the water flow direction. Therefore, by obtaining the installation angle of the flow velocity radar, a reference angle for the water flow direction can be set as a reference standard for correcting the flow velocity data. Subsequently, the flow velocity time series data is traversed using a preset flow velocity threshold, and time periods exceeding or falling below the threshold range are filtered out to locate multiple flow velocity time series segments to be compensated. These time series segments to be compensated correspond to time periods that may be affected by wind speed and direction, and these flow velocity data need to be compensated to improve accuracy. Then, based on the start and end timestamps of these flow velocity time series segments to be compensated, multiple wind speed and direction segments are segmented from the acquired ultrasonic wind speed and direction data. Then, based on the difference between the wind direction and the reference angle of the water flow direction in the wind speed and direction segments, the vector influence of wind speed on the water flow direction is calculated, and this is used to correct the flow velocity time series segments to be compensated, resulting in corrected flow velocity time series segments. Finally, based on the start and end timestamps of the velocity segment to be compensated, the corrected velocity time series segment is smoothly replaced with the velocity time series data to ensure a smooth transition of the data and avoid abrupt or unnatural changes. Through this smooth replacement process, the system obtains corrected velocity data, providing accurate data for further hydrological analysis, flow calculation, and disaster early warning, ensuring the accuracy and reliability of hydrological monitoring results under the interference of wind speed and wind direction.

[0024] Furthermore, the method further includes using the multiple wind speed and direction segments to perform vector compensation calculations on the multiple velocity time series segments to be compensated, thereby obtaining multiple corrected velocity time series segments. Using the reference angle of the water flow direction as a reference zero degree, the angle between the ultrasonic wind speed and wind direction data is calculated, and the wind direction angle time series data is output; according to the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind direction angle time series segments are segmented from the wind direction angle time series data; according to the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind speed time series segments are segmented from the ultrasonic wind speed and wind direction data, wherein the multiple wind direction angle time series segments and the multiple wind speed time series segments constitute the multiple corrected velocity time series segments.

[0025] Optionally, the reference angle for the water flow direction is first set to zero degrees. This reference angle is determined based on the installation direction of the flow velocity radar equipment and represents the main flow direction. Then, real-time wind speed and direction data provided by the ultrasonic anemometer are acquired. For each sampling period, the wind direction angle is subtracted from the reference angle to obtain the wind direction angle. If the angle is negative, 360 degrees are added to ensure it is always positive. The calculated wind direction angle at each moment is the wind direction angle time series data, representing the angular difference between the wind direction and the water flow direction. Next, based on the start and end timestamps of multiple flow velocity time series segments to be compensated, the wind direction angle data for the corresponding time period is extracted from the wind direction angle time series data to form wind direction angle time series segments that match the time range of the flow velocity segments to be compensated. Simultaneously, based on the start and end timestamps of the flow velocity time series segments to be compensated, the corresponding wind speed data is segmented from the ultrasonic anemometer and wind direction data to form multiple wind speed time series segments. Finally, the obtained time series segments of wind direction angle and wind speed are integrated to form multiple corrected flow velocity time series segments. These corrected flow velocity time series segments can effectively eliminate the interference of wind speed and wind direction on the original flow velocity data, provide more accurate flow velocity values, and ensure the accuracy of subsequent hydrological analysis and flow prediction.

[0026] Multiple spatially associated stations of the target monitoring station are linked to retrieve multiple water and rainfall data.

[0027] In one embodiment, a linkage relationship is first established with spatially associated stations related to the target monitoring station to obtain water and rainfall data from these spatially associated stations. Spatially associated stations refer to other monitoring stations geographically distant from the target monitoring station. To ensure data integrity and timeliness, the system sends data requests to each spatially associated station via network communication protocols or data exchange interfaces to retrieve real-time water and rainfall data. This data includes, but is not limited to, water level data, flow velocity data, rainfall data, and meteorological data from each associated station. When retrieving this data, the system uses a set time synchronization mechanism to ensure that the timestamps of the data obtained from different spatially associated stations are consistent, enabling effective comparison and analysis over time. This provides reliable data support for multi-station joint prevention and control flow forecasting, disaster risk analysis, and early warning systems.

[0028] Based on the topographic parameters of the target monitoring station and multiple spatially associated stations, the cross-sectional flow is distributedly corrected on the optimized water and rainfall data and multiple water and rainfall associated data to obtain the multi-station joint prevention flow prediction value.

[0029] In one embodiment, the topographic parameters of the target monitoring station and its multiple spatially associated stations are first acquired. These parameters include the station's topographic features. Comprehensive analysis of these parameters allows for the assessment of the propagation characteristics of water flow between different stations and the trend of flow variation. Subsequently, the system combines the optimized hydrological and rainfall data of the target monitoring station with the associated hydrological and rainfall data retrieved from multiple spatially associated stations to calculate the flow rate of a single station. The calculated single-station flow rate is then loaded into a distributed hydrological network model constructed based on the topographic parameters. The system dynamically activates the calculation nodes in the model according to the river flow direction to correct the node flow, resulting in a more accurate multi-station joint prevention flow prediction. Finally, the system outputs the calculated multi-station joint prevention flow prediction as key data for further disaster early warning and disaster prevention decision support. These predictions provide a comprehensive flow trend for each monitoring station, helping to promptly identify potential risks within the basin, providing a scientific basis for basin management and water resource allocation, and offering data support for downstream disaster risk prediction and emergency response.

[0030] Furthermore, based on the topographic parameters of the target monitoring station and multiple spatially associated stations, the cross-sectional flow is distributedly corrected on the optimized hydrological and rainfall data and multiple associated hydrological and rainfall data to obtain multi-station joint prevention flow prediction values. The method also includes: Based on the optimized hydrological and rainfall data and multiple related hydrological and rainfall data, single-station flow is calculated to obtain the target station baseline flow and multiple related station baseline flows. An RNG κ-ε turbulence-driven distributed hydrological network model is constructed based on the topographic parameters. After loading the target station baseline flow and multiple related station baseline flows into the distributed hydrological network model, the calculation nodes in the distributed hydrological network model are dynamically activated according to the river flow direction to progressively correct the node flow, outputting the distributed joint prevention flow prediction value. The distributed joint prevention flow prediction value is decomposed to obtain the multi-station joint prevention flow prediction value.

[0031] Preferably, based on the obtained optimized hydrological and rainfall data and multiple related hydrological and rainfall data, the baseline flow rate of the target monitoring station and its multiple spatially related stations is calculated using a single-station flow rate calculation method. During the calculation, the corrected velocity, bank coefficient, and cross-sectional area of ​​the river channel at the target station are multiplied to obtain the baseline flow rate at the target station. Similarly, the velocity, bank coefficient, and cross-sectional area of ​​the river channel at the related stations are multiplied to obtain the baseline flow rates at multiple related stations. Subsequently, based on the acquired topographic parameters, an RNG κ-ε turbulence-driven distributed hydrological network model is constructed. Specifically, in the RNG κ-ε turbulence model, κ represents the turbulence intensity of the water flow, and ε represents the dissipation rate of turbulent energy. The turbulence model is mainly used to simulate complex phenomena such as eddies and turbulence in water flow, and can more accurately reflect the motion characteristics of water flow under different topographic and velocity conditions. The system calculates the turbulence energy distribution of each station using the RNG κ-ε model based on the topographic parameters within the watershed. During model construction, based on topographic topology and flow characteristics, the flow and turbulence parameters of each computational node are set. Each computational node represents a hydrological station. By solving the turbulence dynamics equations, the turbulent kinetic energy and dissipation rate of each station are obtained. As the water propagates, these turbulence parameters affect the flow and flow of downstream nodes. Therefore, the system dynamically activates each node in the hydrological network through the water flow transport equations, gradually correcting the flow data of the nodes according to the watershed's water flow propagation path and turbulence effects. When loading the target station's baseline flow and the baseline flows of multiple related stations into the constructed distributed hydrological network model, the system calculates the flow of the nodes step by step from upstream to downstream, following the river flow direction. At each node, the flow, turbulence intensity, dissipation rate, and topographic parameters of the previous node are combined to calculate the flow of that node. This process is carried out iteratively until the flow of all nodes reaches a convergence state, i.e., the magnitude of flow change is less than a preset threshold. Once the flow of all nodes has been corrected, the model outputs the distributed joint prevention and control flow prediction value. Finally, the system decomposes the generated distributed joint defense traffic prediction value. That is, according to the location of each computing node in the model, the multi-station joint defense traffic prediction value is broken down into the traffic prediction value of each station, thus forming the multi-station joint defense traffic prediction value, which provides a reliable basis for subsequent disaster risk assessment and emergency response.

[0032] Based on the multi-station joint prevention traffic prediction value, the scenario disaster risk is predicted, and a disaster warning command is output, wherein the disaster warning command carries a disaster warning range identifier.

[0033] In one embodiment, the predicted flow rate of the multi-station joint prevention and control system is first converted using the relationship between flow rate and water level to obtain the predicted water level values ​​for each monitoring station. These predicted water level values ​​represent the possible changes in water level over a future period, serving as the basis for subsequent risk prediction. Subsequently, time difference processing is performed on multiple predicted water level values ​​to calculate the water level change rate. This rate reflects the speed of water level change; a large rate indicates a possible sudden rise or fall in water level, leading to an increased risk of disaster. Simultaneously, the system retrieves radar rainfall time-series data and rainfall time-series data from multiple associated water and rainfall data sources from optimized water and rainfall data and multiple associated water and rainfall data sources, forming multiple scenario rainfall time-series data. Then, sliding window extreme value extraction is performed on these scenario rainfall time-series data to obtain multiple rainfall intensity extreme values. Using preset water level mutation thresholds and rainstorm thresholds as dual trigger conditions, the calculated water level change rate and rainfall intensity extreme values ​​are iterated through to select P risk stations. These risk stations are areas with drastic water level changes and extremely high rainfall intensity, potentially triggering floods. Next, these risk stations were connected according to the river's flow direction, and river buffer zones were overlaid to construct a pre-disaster warning range. This range identifies the geographical extent of potentially affected areas, helping relevant departments clarify the areas that may be impacted by the disaster. Finally, based on the water level change rate and extreme rainfall intensity of these risk stations, the warning level for each risk station was quantified, and a pre-disaster warning instruction was generated. This instruction includes the pre-disaster warning range identifier, helping relevant departments to take timely and effective measures to reduce disaster losses.

[0034] Furthermore, based on the multi-station joint prevention traffic prediction values, scenario-based disaster risk prediction is performed, and a disaster warning command is output. The disaster warning command includes a disaster warning range identifier. The method also includes: The multi-station joint prevention flow prediction values ​​are converted into multiple predicted water level values; time difference processing is performed on the multiple predicted water level values ​​to calculate multiple water level change rates; multiple radar rainfall time series data and multiple associated rainfall time series data are called from the optimized water and rainfall data and multiple water and rainfall correlation data respectively to form multiple scene rainfall time series data; sliding window extreme value extraction is performed on the multiple scene rainfall time series data to obtain multiple rainfall intensity extreme values; using preset water level change threshold and rainstorm threshold as dual trigger conditions, the multiple water level change rates and multiple rainfall intensity extreme values ​​are traversed to perform spatial topological correlation narrowing and filter out P risk stations; after connecting the P risk stations according to the river flow direction, a river buffer is superimposed to construct the disaster warning range; the warning level is quantified based on the P water level change rates and P rainfall intensity extreme values ​​of the P risk stations, and the disaster warning command is generated.

[0035] Preferably, the system first performs an inverse operation on the aforementioned baseline flow rate to obtain multiple predicted water level values. Each predicted water level value represents the water level change based on the existing flow rate prediction over a future period, providing fundamental data for subsequent risk prediction. Then, time-difference processing is performed on the multiple predicted water level values ​​to calculate the water level change rate. The water level change rate represents the rate of change of water level per unit time, reflecting the trend and speed of water level change. A large water level change rate indicates drastic water level changes, which may lead to floods or other hydrological disasters. Time-difference allows for the quantification of the magnitude of water level change at each time step, providing crucial information for subsequent risk analysis. Simultaneously, the system retrieves multiple radar rainfall time-series data and multiple associated rainfall time-series data from optimized hydrological and rainfall data and multiple hydrological and rainfall correlation data, forming multiple scenario rainfall time-series data. The radar rainfall data and associated rainfall data reflect rainfall conditions at different times and locations. The system combines these time-series data into multiple scenario rainfall data, comprehensively reflecting the rainfall intensity and distribution at different locations within the region, providing multi-dimensional rainfall information for disaster early warning. For these scenario rainfall time-series data, the system performs sliding window extreme value extraction. In this process, the sliding window method extracts the extreme values ​​of rainfall intensity within a certain time window, i.e., the maximum rainfall amount. These extreme values ​​represent the maximum rainfall intensity in each time period and are important criteria for determining whether rainfall reaches the rainstorm standard and whether it may trigger hydrological disasters. Through extreme value extraction, multiple rainfall intensity extreme values ​​can be obtained, and the potential for future extreme rainfall events can be assessed. Then, using preset water level change thresholds and rainstorm thresholds as dual trigger conditions, the system iterates through the calculated water level change rate and rainfall intensity extreme values. When a water level change rate exceeds the preset change threshold, or a rainfall intensity exceeds the rainstorm threshold, the station is considered to have potential disaster risk. Based on this, the system uses spatial topological correlation narrowing to select P risk stations. Spatial topological correlation narrowing refers to determining which stations are more affected by adjacent areas based on the water flow transmission path, and selecting these stations as high-risk areas. After identifying P risk stations, these stations are connected based on the river's flow direction. Then, a river buffer zone is overlaid to construct a pre-disaster warning area. This buffer zone refers to the area within a preset distance around the river that may be affected by rising water levels or flooding. By overlaying the buffer zone, the pre-disaster warning area can be clearly defined, providing a more precise picture of the disaster-affected area. Finally, based on the water level change rate and extreme rainfall intensity of the P risk stations, the warning level is quantified using a lookup table. This warning level reflects the severity of the hydrological disaster affecting that station. Based on the quantified warning level, the system generates corresponding pre-disaster warning instructions to help relevant departments take timely countermeasures and reduce disaster losses.

[0036] In summary, the embodiments of this application have at least the following technical effects: First, a time-domain accumulation algorithm is used for real-time monitoring of low signal-to-noise ratio scenarios involving radar echo energy at the target monitoring station. Based on the monitoring results, radar signal anti-interference enhancement is performed, and time-series water and rainfall data is output. Next, the time-series water and rainfall data is directionally corrected based on ultrasonic wind speed and direction data, resulting in optimized water and rainfall data. Then, multiple spatially associated stations of the target monitoring station are linked, retrieving multiple water and rainfall correlation data. Next, based on the terrain topology parameters of the target monitoring station and the multiple spatially associated stations, distributed cross-sectional flow correction is performed on the optimized water and rainfall data and the multiple water and rainfall correlation data to obtain multi-station joint prevention flow prediction values. Finally, based on the multi-station joint prevention flow prediction values, scenario-based disaster risk prediction is performed, and a disaster warning command is output, which includes a disaster warning range identifier. It solves the technical problems of low accuracy and poor data consistency in water and rainfall monitoring under low signal-to-noise ratio, complex terrain and strong interference environments, making it difficult to achieve multi-site collaborative flow prediction. It achieves the technical effect of improving the quality and stability of water and rainfall data under low signal-to-noise ratio environments through multi-source sensor data fusion and multi-station joint prevention and collaborative correction, and realizes high-precision distributed prediction of watershed cross-sectional flow.

[0037] Example 2, based on the same inventive concept as the microwave radar-based integrated water and rainfall sensing optimization method in the previous examples, such as... Figure 2 As shown, this application provides an integrated water and rainfall information sensing and optimization system based on microwave radar. The system includes: Radar signal detection module 11: Employs a time-domain accumulation algorithm to monitor low signal-to-noise ratio scenarios of water and rainfall radar echo energy at the target monitoring station in real time, and enhances radar signal anti-interference based on the monitoring results, outputting water and rainfall time-series data; Directional correction module 12: Performs directional correction on the water and rainfall time-series data based on ultrasonic wind speed and direction data, outputting optimized water and rainfall data; Data retrieval module 13: Links multiple spatially associated stations of the target monitoring station to retrieve multiple water and rainfall associated data; Cross-sectional flow correction module 14: Performs distributed cross-sectional flow correction on the optimized water and rainfall data and multiple water and rainfall associated data based on the terrain topology parameters of the target monitoring station and multiple spatially associated stations, obtaining multi-station joint prevention flow prediction values; Risk prediction module 15: Performs scenario-based disaster risk prediction based on the multi-station joint prevention flow prediction values, outputting a disaster warning command, wherein the disaster warning command carries a disaster warning range identifier.

[0038] Furthermore, the orientation correction module 12 is used to perform the following method: Flow velocity time-series data is extracted from the water and rainfall time-series data, which includes water level time-series data, flow velocity time-series data, and radar rainfall time-series data. Wind disturbance error compensation is performed on the flow velocity time-series data based on the ultrasonic wind speed and direction data to obtain corrected flow velocity data. The corrected flow velocity data, water level time-series data, and radar rainfall time-series data constitute optimized water and rainfall data.

[0039] Furthermore, the orientation correction module 12 is used to perform the following method: The first echo energy value of the water level radar is monitored in real time. If the first echo energy value is lower than a preset first threshold, a low signal-to-noise ratio water level monitoring scenario is determined. In the low signal-to-noise ratio water level monitoring scenario, the sampling time delay of the water level radar is applied, and an extended sampling time window is set. The echo signal sequence of the water level radar is subjected to superposition and moving average filtering processing in the extended sampling time window, and the water level time series data is output.

[0040] Furthermore, the orientation correction module 12 is used to perform the following method: Based on the installation direction of the flow velocity radar equipment at the target monitoring station, the reference angle of the water flow direction is determined; the flow velocity time series data is traversed using a preset flow velocity threshold to locate multiple flow velocity time series segments to be compensated; based on the start and end timestamps of the multiple flow velocity time series segments to be compensated, multiple wind speed and direction segments are segmented from the ultrasonic wind speed and direction data; vector compensation calculations are performed on the multiple flow velocity time series segments to be compensated using the multiple wind speed and direction segments to obtain multiple corrected flow velocity time series segments; based on the start and end timestamps of the multiple flow velocity time series segments to be compensated, the multiple corrected flow velocity time series segments are smoothly replaced in the flow velocity time series data to complete wind disturbance error compensation and obtain the corrected flow velocity data.

[0041] Furthermore, the orientation correction module 12 is used to perform the following method: Using the reference angle of the water flow direction as a reference zero degree, the angle between the ultrasonic wind speed and wind direction data is calculated, and the wind direction angle time series data is output; according to the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind direction angle time series segments are segmented from the wind direction angle time series data; according to the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind speed time series segments are segmented from the ultrasonic wind speed and wind direction data, wherein the multiple wind direction angle time series segments and the multiple wind speed time series segments constitute the multiple corrected velocity time series segments.

[0042] Furthermore, the cross-sectional flow correction module 14 is used to perform the following method: Based on the optimized hydrological and rainfall data and multiple related hydrological and rainfall data, single-station flow is calculated to obtain the target station baseline flow and multiple related station baseline flows. An RNG κ-ε turbulence-driven distributed hydrological network model is constructed based on the topographic parameters. After loading the target station baseline flow and multiple related station baseline flows into the distributed hydrological network model, the calculation nodes in the distributed hydrological network model are dynamically activated according to the river flow direction to progressively correct the node flow, outputting the distributed joint prevention flow prediction value. The distributed joint prevention flow prediction value is decomposed to obtain the multi-station joint prevention flow prediction value.

[0043] Furthermore, the risk prediction module 15 is used to perform the following method: The multi-station joint prevention flow prediction values ​​are converted into multiple predicted water level values; time difference processing is performed on the multiple predicted water level values ​​to calculate multiple water level change rates; multiple radar rainfall time series data and multiple associated rainfall time series data are called from the optimized water and rainfall data and multiple water and rainfall correlation data respectively to form multiple scene rainfall time series data; sliding window extreme value extraction is performed on the multiple scene rainfall time series data to obtain multiple rainfall intensity extreme values; using preset water level change threshold and rainstorm threshold as dual trigger conditions, the multiple water level change rates and multiple rainfall intensity extreme values ​​are traversed to perform spatial topological correlation narrowing and filter out P risk stations; after connecting the P risk stations according to the river flow direction, a river buffer is superimposed to construct the disaster warning range; the warning level is quantified based on the P water level change rates and P rainfall intensity extreme values ​​of the P risk stations, and the disaster warning command is generated.

[0044] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for integrated water and rainfall sensing optimization based on microwave radar, characterized in that, The method includes: A time-domain accumulation algorithm is used to monitor the low signal-to-noise ratio scenario of water and rainfall radar echo energy at the target monitoring station in real time, and the radar signal anti-interference enhancement is performed based on the monitoring results to output water and rainfall time series data. Based on ultrasonic wind speed and direction data, the water and rainfall time series data are directionally corrected to output optimized water and rainfall data. Link multiple spatially associated stations of the target monitoring station to retrieve multiple water and rainfall data; Based on the topographic parameters of the target monitoring station and multiple spatially associated stations, the cross-sectional flow distributed correction is performed on the optimized water and rainfall data and multiple water and rainfall associated data to obtain the multi-station joint prevention flow prediction value. Based on the multi-station joint prevention traffic prediction value, the scenario disaster risk is predicted, and a disaster warning command is output, wherein the disaster warning command carries a disaster warning range identifier. The method further includes: Based on the topographic parameters of the target monitoring station and multiple spatially associated stations, performing distributed cross-sectional flow correction on the optimized hydrological and rainfall data and multiple associated hydrological and rainfall data to obtain multi-station joint prevention flow prediction values; the method also includes: Based on the optimized hydrological and rainfall data and multiple related hydrological and rainfall data, single-station flow is calculated to obtain the target station baseline flow and multiple related station baseline flows. An RNG κ-ε turbulence-driven distributed hydrological network model was constructed based on the aforementioned terrain topology parameters. After loading the target station's baseline flow and the baseline flows of multiple associated stations into the distributed hydrological network model, the computing nodes in the distributed hydrological network model are dynamically activated according to the river flow direction to progressively correct the node flow and output the distributed joint defense flow prediction value. The distributed joint defense traffic prediction value is decomposed to obtain the multi-station joint defense traffic prediction value; The method includes: predicting the scenario's impending disaster risk based on the multi-station joint prevention traffic prediction values, and outputting an impending disaster warning command, wherein the impending disaster warning command carries an impending disaster warning range identifier; the method further includes: The predicted flow rate from the multi-station joint prevention and control system is converted into multiple predicted water level values. The predicted water level values ​​are subjected to time difference processing to calculate multiple water level change rates; Multiple radar rainfall time series data and multiple associated rainfall time series data are called from the aforementioned water and rainfall optimization data and multiple water and rainfall correlation data respectively to form multiple scene rainfall time series data; Multiple rainfall intensity extreme values ​​were obtained by performing sliding window extreme value extraction on the rainfall time series data of the multiple scenarios. Using preset water level change thresholds and rainstorm thresholds as dual triggering conditions, the system iterates through multiple water level change rates and multiple extreme rainfall intensity values, performs spatial topological correlation narrowing, and filters out P risk sites. After connecting the P risk stations according to the river flow direction, the river buffer zone is superimposed to construct the disaster early warning range. The pre-disaster warning command is generated based on the P water level change rates and P extreme rainfall intensity values ​​of the P risk stations.

2. The integrated water and rainfall sensing optimization method based on microwave radar as described in claim 1, characterized in that, The method further includes: Flow velocity time-series data are extracted from the water and rainfall time-series data, wherein the water and rainfall time-series data includes water level time-series data, flow velocity time-series data and radar rainfall time-series data; Based on the ultrasonic wind speed and direction data, wind disturbance error compensation is performed on the flow velocity time series data to obtain corrected flow velocity data. The corrected flow velocity data, water level time series data, and radar rainfall time series data constitute optimized water and rainfall data.

3. The integrated water and rainfall sensing optimization method based on microwave radar as described in claim 2, characterized in that, The method further includes: The first echo energy value of the water level radar is monitored in real time. If the first echo energy value is lower than a preset first threshold, a low signal-to-noise ratio water level monitoring scenario is determined. In the low signal-to-noise ratio water level monitoring scenario, the sampling delay of the water level radar is adjusted, and an extended sampling time window is set. The echo signal sequence of the water level radar is superimposed and subjected to moving average filtering within the extended sampling time window to output the water level time series data.

4. The integrated water and rainfall sensing optimization method based on microwave radar as described in claim 2, characterized in that, Based on the ultrasonic wind speed and direction data, wind disturbance error compensation is performed on the flow velocity time series data to obtain corrected flow velocity data. The method further includes: The reference angle for the water flow direction is determined based on the installation direction of the flow velocity radar equipment at the target monitoring station; The flow velocity time series data is traversed using a preset flow velocity threshold to locate multiple flow velocity time series segments that need to be compensated. Based on the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind speed and direction segments are segmented from the ultrasonic wind speed and direction data. Vector compensation calculations are performed on the multiple wind speed and direction segments to obtain multiple corrected flow velocity time series segments. Based on the start and end timestamps of the multiple velocity time series segments to be compensated, the multiple corrected velocity time series segments are smoothly replaced in the velocity time series data to complete wind disturbance error compensation and obtain the corrected velocity data.

5. The integrated water and rainfall sensing optimization method based on microwave radar as described in claim 4, characterized in that, The method further includes: using the multiple wind speed and direction segments to perform vector compensation calculations on the multiple time series segments of the flow velocity to be compensated, thereby obtaining multiple corrected flow velocity time series segments; Using the reference angle of the water flow direction as the reference zero degree, the angle between the ultrasonic wind speed and wind direction data is calculated, and the wind direction angle time series data is output. Based on the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind direction angle time series segments are segmented from the wind direction angle time series data; Based on the start and end timestamps of the multiple velocity time series segments to be compensated, multiple wind speed time series segments are segmented from the ultrasonic wind speed and wind direction data, wherein the multiple wind direction angle time series segments and the multiple wind speed time series segments constitute the multiple corrected velocity time series segments.

6. A microwave radar-based integrated water and rainfall sensing optimization system, characterized in that, The system is used to implement the integrated water and rainfall information sensing and optimization method based on microwave radar as described in any one of claims 1-5, the system comprising: Radar signal detection module: It adopts a time-domain accumulation algorithm to monitor the low signal-to-noise ratio scenario of water and rainfall radar echo energy at the target monitoring station in real time, and enhances the anti-interference of radar signals based on the monitoring results, and outputs water and rainfall time series data. Targeted correction module: Based on ultrasonic wind speed and direction data, the water and rainfall time series data are targeted and corrected, and optimized water and rainfall data are output. Data retrieval module: Links multiple spatially associated stations of the target monitoring station to retrieve multiple water and rainfall data; Cross-sectional flow correction module: Based on the topographic parameters of the target monitoring station and multiple spatially associated stations, the cross-sectional flow is distributedly corrected on the optimized water and rainfall data and multiple water and rainfall associated data to obtain the multi-station joint prevention flow prediction value; Risk prediction module: Based on the multi-station joint prevention traffic prediction value, it performs scenario-based disaster risk prediction and outputs a disaster warning command, wherein the disaster warning command carries a disaster warning range identifier; The cross-sectional flow correction module further includes: calculating the single-station flow based on the optimized hydrological and rainfall data and multiple related hydrological and rainfall data to obtain the target station baseline flow and multiple related station baseline flows; constructing an RNG κ-ε turbulence-driven distributed hydrological network model based on the topographic parameters; loading the target station baseline flow and multiple related station baseline flows into the distributed hydrological network model, and dynamically activating the calculation nodes in the distributed hydrological network model according to the river flow direction to progressively correct the node flow, outputting the distributed joint prevention flow prediction value; and decomposing the distributed joint prevention flow prediction value to obtain the multi-station joint prevention flow prediction value. The risk prediction module further includes: converting the multi-station joint prevention flow prediction value into multiple predicted water level values; performing time difference processing on the multiple predicted water level values ​​to calculate multiple water level change rates; calling multiple radar rainfall time series data and multiple associated rainfall time series data from the optimized water and rainfall data and multiple water and rainfall correlation data respectively to form multiple scenario rainfall time series data; performing sliding window extreme value extraction on the multiple scenario rainfall time series data to obtain multiple rainfall intensity extreme values; using preset water level change thresholds and rainstorm thresholds as dual triggering conditions, traversing the multiple water level change rates and multiple rainfall intensity extreme values, performing spatial topological correlation narrowing, and filtering out P risk stations; connecting the P risk stations based on the river flow direction sequence, and superimposing a river buffer to construct the disaster early warning range; quantifying the early warning level based on the P water level change rates and P rainfall intensity extreme values ​​of the P risk stations, and generating the disaster early warning command.

Images