Artificial precipitation enhancement operation effect evaluation method based on index threshold value method

Abstract

The invention relates to an artificial precipitation enhancement operation effect evaluation method based on an index threshold method. The method comprises the following steps: fusing multi-source data to construct a standardized spatio-temporal data cube; an objective space region candidate set is generated through unsupervised clustering; in combination with a Lagrange particle diffusion model and dynamic feature similarity matching, a target area and a contrast area are automatically and optimally determined from the candidate set, and subjective bias is eliminated from the source; extracting and calculating a net effect difference value sequence of the key indexes of the two regions; and generating a quantitative evaluation conclusion through integrated calculation and comparison with a statistical threshold. According to the method, objectification and self-adaption of comparison area selection can be realized, the uncertainty of subjective selection can be quantified, and the accuracy and credibility of effect evaluation under complex terrain and weather conditions are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of weather modification and meteorological data processing technology, specifically to a method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method. Background Technology

[0002] Artificial rain enhancement, as an important technical means to alleviate water shortages and cope with specific meteorological disasters, has always had a critical issue in the scientific evaluation of its operational effectiveness. The indicator threshold method is a widely used approach in current operational practice. Its core principle lies in comparing the changes in key meteorological indicators such as precipitation and radar reflectivity before and after the operation between the artificially affected target area and the unaffected control area. This difference is then compared to a statistical threshold representing natural fluctuations to determine whether the operation has produced a significant effect exceeding natural variations. This type of method has become the mainstream technical approach for evaluating operational effectiveness due to its relatively clear logic and strong operability.

[0003] However, the traditional index threshold method suffers from a fundamental technical bottleneck in practical applications: the objective selection of the comparison zone. Existing technologies typically rely on the experience of operators, manually delineating a static comparison zone upwind or crosswind of the target area. While this method may be feasible in simple weather conditions and flat terrain, its shortcomings become glaringly apparent in complex mountainous terrain or rapidly evolving weather systems. First, the manually selected comparison zone and the target area cannot guarantee true similarity in terms of climate background and topographic lifting, leading to different "background responses" to natural weather processes and introducing significant systematic bias or "noise" into the assessment results. Second, the static and fixed delineation method cannot adapt to the unique wind fields and cloud structures of each operation, potentially resulting in the comparison zone being "contaminated" by catalysts or operating under entirely different cloud physics processes, rendering the assessment meaningless. Finally, because the selection process is highly dependent on subjective experience, different personnel or different approaches can lead to vastly different or even contradictory assessment conclusions. Furthermore, this uncertainty introduced by "human subjectivity" cannot be quantitatively measured or reported, severely undermining the reliability and credibility of the assessment results. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide an evaluation method for artificial rain enhancement operations based on the index threshold method that can automatically, objectively, and optimally determine the comparison area and quantify the uncertainty of subjective selection during the evaluation process.

[0005] The objective of this invention is achieved through the following solution:

[0006] In a first aspect, the present invention provides a method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method, comprising the following steps:

[0007] S1: Standardize and fuse the raw data, which includes the geographical background of the work area, real-time meteorological observations and numerical forecasts, and align the raw data with spatiotemporal references and unify the format to generate a standardized spatiotemporal data cube covering the entire region.

[0008] S2: Perform multi-dimensional feature extraction on the spatiotemporal data cube, and automatically classify the grid points based on their climate background, terrain and real-time meteorological features to generate multiple candidate sets of spatial regions with similar internal features that can serve as potential analysis units.

[0009] S3: Based on wind field data in the spatiotemporal data cube, the transport trajectory of the catalyst in the target cloud system is simulated by the Lagrange particle diffusion model. The region with the highest overlap with the simulated diffusion path is selected from the candidate set of spatial regions to generate the optimal target area.

[0010] S4: In the candidate set of spatial regions, calculate and select another region with the highest comprehensive similarity to the optimal target region in terms of climate background and real-time meteorological characteristics, and generate the optimal comparison region;

[0011] S5: Based on the optimal target area and the optimal comparison area, extract the precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube, calculate the index time series of the optimal target area and the optimal comparison area respectively, and perform background value subtraction to generate the catalytic effect difference time series characterizing the net effect of the catalytic operation.

[0012] S6: Perform integrated calculations on the time series of catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate an evaluation conclusion on the operation effect.

[0013] In one embodiment, S1 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0014] S11: Perform gridded interpolation on the static geographic background data in the raw data containing the geographic background of the work area, real-time meteorological observations and numerical forecasts, and convert the discrete geographic information data into a global continuous background field with the same spatial resolution as the meteorological observation data, generating a standardized static background data layer.

[0015] S12: Perform spatiotemporal alignment and quality control processing on the real-time meteorological observation and numerical forecast data in the raw data, remove outliers, and uniformly interpolate data from different sources and at different times to the same spatiotemporal grid point to generate a standardized dynamic meteorological data layer.

[0016] S13: Perform multi-dimensional fusion processing on the static background data layer and the dynamic meteorological data layer. According to the grid coordinates and time series, the two types of data are superimposed and integrated into a multi-dimensional array containing geography, climate and real-time meteorological status, generating a standardized spatiotemporal data cube for the entire domain.

[0017] In one embodiment, S2 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0018] S21: Perform feature dimension traversal and calculation processing on the spatiotemporal data cube, extract feature values ​​of multiple dimensions such as elevation, historical precipitation climate value, land surface type, and reflectivity, brightness temperature, and wind field at the start of the operation for each grid point, and generate a comprehensive feature vector for each grid point.

[0019] S22: Perform unsupervised classification based on density clustering on the comprehensive feature vector of all grid points. Automatically identify and merge grid points with similar features according to the distribution density of feature vectors in multidimensional space, and generate multiple spatial grid point clusters with high internal homogeneity.

[0020] S23: Perform geographic connectivity and meteorological rationality verification on the spatial grid clusters, eliminate discrete noise points, and encapsulate the spatial range vectorization of the grid clusters that meet the verification to generate a candidate set of spatial regions.

[0021] In one embodiment, S3 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0022] S31: Perform vertical integration and smoothing on the multi-level wind field data in the spatiotemporal data cube, calculate the guiding airflow field representing the average catalyst delivery path, and generate the driving wind field for particle diffusion simulation.

[0023] S32: Based on the driving wind field, the Lagrange particle reverse trajectory simulation is performed. A large number of virtual particles are released from the operation point and the three-dimensional backward trajectory and spatial residence probability of the virtual particles within a certain period of time are calculated based on the wind field data to generate the potential influence range probability field of the catalyst.

[0024] S33: Perform spatial overlay and overlap analysis on the probability field of the potential influence range and the candidate set of spatial regions. Calculate the spatial fit between each candidate region and the high-value area of ​​the probability field of the potential influence range. The candidate region with the highest spatial fit is determined as the optimal target region. The formula for calculating the spatial fit is:

[0025]

[0026] in, Candidate region Space adaptability, For grid points The probability value of the catalyst at that location. This refers to the high-probability region comprised of all grid points in the probability field whose probability values ​​exceed a preset probability threshold. Spatial weights related to the distance from the grid point to the work point and the terrain uplift effect, Candidate region The total number of grid points contained.

[0027] In one embodiment, S4 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0028] S41: After excluding the optimal target area from the spatial region candidate set, representative feature extraction processing is performed on the remaining candidate areas. The static feature mean vector and the change sequence of dynamic meteorological features before and after the operation are calculated for each area to generate a standardized feature profile of each candidate area.

[0029] S42: Perform multi-dimensional similarity calculations between the standardized feature profiles of each candidate region and the corresponding features of the optimal target region. Comprehensively evaluate the degree of matching between the two in terms of climate background, topographic attributes, and the temporal evolution of key meteorological indicators. Generate a comprehensive similarity score for each candidate region relative to the target region. The formula for calculating the comprehensive similarity score is as follows:

[0030]

[0031] in, The score represents the combined similarity between candidate region i and the optimal target region T. and These are the static background feature vectors of the target region T and the candidate region i, respectively. Let be the variance of the distances between the static feature vectors of all candidate regions. and These are time series of key dynamic meteorological indicators for the target area T and candidate area i within a preset time period before the operation. For all candidate regions and target region T The maximum distance, and Weighting coefficients to adjust the contributions of static and dynamic similarity;

[0032] S43: Compare, sort, and filter the comprehensive similarity scores of all candidate regions, select the candidate region with the highest score that exceeds the preset similarity threshold, and generate the optimal comparison region.

[0033] In one embodiment, S5 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0034] S51: Based on the spatial boundaries of the optimal target area and the optimal comparison area, spatial mask extraction processing is performed on the spatiotemporal data cube. Precipitation rate, radar combined reflectivity and other index data of all grid points in the two areas during the baseline period before operation and the evaluation period after operation are extracted respectively to generate the original spatiotemporal dataset of the optimal target area and the optimal comparison area.

[0035] S52: Perform time series aggregation on the original spatiotemporal dataset of indicators, calculate the average indicator values ​​of the optimal target area and the optimal comparison area at each evaluation time, and remove the average state value of each time period before the baseline period before the operation from the value of each time period to generate the indicator increment sequence after background subtraction of the optimal target area and the optimal comparison area.

[0036] S53: Perform time-by-time difference processing on the index increment sequence, calculate the difference between the index increment of the optimal target area and the index increment of the optimal comparison area at the same time, and generate a time series of catalytic effect difference that characterizes the net effect of catalytic operation.

[0037] In one embodiment, step S6 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0038] S61: Perform time-weighted integral processing based on physical processes on the time series of catalytic effect difference, assign different contribution weights to the values ​​at different times in the series according to the typical time period characteristics of the occurrence and development of catalysis, and accumulate the weighted values ​​to generate the integrated effect value.

[0039] S62: Perform statistical comparative analysis on the integrated effect value and the preset significance threshold to determine whether the integrated effect value exceeds the critical level defined by the preset significance threshold, and generate a qualitative judgment result on whether the operation effect is significant.

[0040] S63: Based on the qualitative judgment results, organize and output the conclusions, integrate the qualitative judgment results and the integrated effect value into a structured evaluation statement, and generate the operation effect evaluation conclusion. The operation effect evaluation conclusion is used to indicate whether the artificial rain enhancement operation has produced a statistically significant catalytic effect.

[0041] Secondly, the present invention provides an evaluation system for the effectiveness of artificial rain enhancement operations based on the index threshold method, the system being configured with the following modules:

[0042] The spatiotemporal data standardization and fusion module is used to standardize and fuse raw data that includes the geographical background of the work area, real-time meteorological observations and numerical forecasts. It aligns the raw data with spatiotemporal references and unifies the format to generate a standardized spatiotemporal data cube covering the entire domain.

[0043] The multidimensional feature region candidate generation module is used to extract multidimensional features from the spatiotemporal data cube, automatically classify the grid points based on the climate background, terrain and real-time meteorological features, and generate multiple spatial region candidate sets with similar internal features that can be used as potential analysis units.

[0044] The optimal target area selection module is used to simulate the transport trajectory of the catalyst in the target cloud system based on wind field data in the spatiotemporal data cube and the Lagrange particle diffusion model. It selects the area with the highest overlap with the simulated diffusion path from the candidate set of spatial regions to generate the optimal target area.

[0045] The optimal comparison area selection module is used to calculate and select the other region with the highest comprehensive similarity to the optimal target region in terms of climate background and real-time meteorological characteristics from the spatial region candidate set, and generate the optimal comparison area.

[0046] The catalytic effect difference sequence generation module is used to extract precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube based on the optimal target area and the optimal comparison area. It calculates the index time series of the optimal target area and the optimal comparison area respectively and performs background value subtraction to generate the catalytic effect difference time series characterizing the net effect of the catalytic operation.

[0047] The task effectiveness evaluation module is used to perform integrated calculations on the time series of catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate the task effectiveness evaluation conclusion.

[0048] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the above-mentioned methods for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method.

[0049] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned methods for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method.

[0050] In summary, the artificial rain enhancement operation effect evaluation method based on the index threshold method provided in this application can completely replace the subjective operation of manually delineating areas based on experience by constructing a standardized spatiotemporal data cube and an objective region candidate set, eliminating the systematic bias introduced by human selection from the data source. By integrating the physical simulation of the Lagrange particle diffusion model with multi-dimensional feature dynamic matching, it can automatically and adaptively determine the optimal target area and comparison area with clear physical meaning and comparable climatic background according to the specific weather conditions and geographical environment of each operation, effectively overcoming the shortcomings of traditional methods in terms of poor applicability under complex terrain and rapidly evolving weather systems. Based on this, the method can achieve precise stripping and quantification of the net effect of catalytic operations, and generate quantitative and reliable operation effect evaluation conclusions by objectively comparing with statistical significance thresholds. Ultimately, this automated process not only significantly improves the objectivity and accuracy of the evaluation results, but also inherently contains the ability to analyze the uncertainty of the evaluation conclusions, providing solid and reliable technical support for the scientific evaluation and decision-making of artificial rain enhancement operations.

[0051] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0052] Figure 1 A flowchart illustrating an evaluation method for artificial rain enhancement operations based on the index threshold method provided in this application embodiment;

[0053] Figure 2 This is a schematic diagram of the structure of an artificial rain enhancement operation effect evaluation system based on the index threshold method, provided in another embodiment of this application. Detailed Implementation

[0054] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0056] In one embodiment, such as Figure 1As shown, a method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and to a system including both a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0057] S1: Standardize and fuse the raw data, which includes the geographical background of the work area, real-time meteorological observations and numerical forecasts, and align the raw data with spatiotemporal references and unify the format to generate a standardized spatiotemporal data cube covering the entire region.

[0058] Specifically, the system collects geographic background data, real-time meteorological observation data, and numerical forecast data for the operational area. The geographic background data for the operational area includes topographic elevation data, landform type data, and administrative boundary data of the operational area and its surroundings, sourced from relevant geographic data released by the National Geographic Information System. The real-time meteorological observation data includes precipitation-related observation data and meteorological element observation data from ground meteorological stations in and around the operational area, radar reflectivity-related observation data and radial velocity observation data from radar stations, and cloud top-related observation data from satellite cloud images, sourced from the national meteorological observation network, weather radar network, and geostationary meteorological satellites. The numerical forecast data includes wind field data, cloud microphysical parameter data, and precipitation forecast data output from numerical weather prediction models, sourced from global or regional numerical weather prediction models.

[0059] Furthermore, the system performs data cleaning on the collected raw data, removing outliers and missing data. Missing data is filled using nearest-neighbor interpolation. Simultaneously, the time resolution of different data sources is uniformly adjusted to ensure consistency across all data in the time dimension. Subsequently, the system performs spatiotemporal benchmark alignment. In spatial benchmark alignment, the system selects the geographic coordinate system of the work area as a unified benchmark and performs coordinate transformation on all raw data to ensure precise spatial matching of all data. In temporal benchmark alignment, the system unifies the timestamps of all data to the same time standard and eliminates time errors caused by different observation equipment through a timestamp calibration algorithm, ensuring strict synchronization of the time series of all data. After completing the spatiotemporal benchmark alignment, the system encapsulates the preprocessed spatiotemporal data using a unified structured data format, constructing a global standardized spatiotemporal data cube. This cube has a three-dimensional structure, comprising spatial, temporal, and indicator dimensions. The spatial dimension uses latitude and longitude grids as basic units, covering the entire operational area and its surrounding area. The temporal dimension covers key periods from before to after the operation, using a unified time step. The indicator dimension encompasses various core meteorological indicators such as precipitation, cloud physics, wind field, and radar reflectivity factors.

[0060] S2: Perform multi-dimensional feature extraction on the spatiotemporal data cube, and automatically classify it based on the climate background, terrain and real-time meteorological features of the grid points to generate multiple candidate sets of spatial regions with similar internal features that can be used as potential analysis units.

[0061] Specifically, for each latitude and longitude grid point in the spatiotemporal data cube, the system extracts three types of core feature parameters and constructs a feature vector. These three types of core feature parameters are climate background features, topographic features, and real-time meteorological features. Climate background features consist of various climate statistical parameters of the grid point over a continuous period prior to the operation, characterizing the long-term climate background of the region. Topographic features consist of various topographic-related parameters of the grid point itself, characterizing the dynamic lifting effect of topography on airflow. Real-time meteorological features consist of various real-time meteorological-related parameters of the grid point during the operation, characterizing the real-time cloud and wind field characteristics of the region. The system performs standardization processing on the feature vector constructed for each grid point, mapping the numerical range of the feature vector to a unified interval, eliminating the influence of differences in the dimensions of different feature parameters, and ensuring that all types of feature parameters have equal weight in subsequent cluster analysis.

[0062] Furthermore, the system can employ clustering algorithms to perform cluster analysis on the feature vectors of all grid points. The system first determines the relevant parameters for clustering analysis, using the spatial distance between grid points as the initial clustering distance, the similarity of feature vectors as the metric, and setting the minimum sample size for each cluster to ensure that the region size corresponding to each cluster meets the requirements of subsequent analysis. Subsequently, the system performs a bottom-up iterative clustering process on all grid points using the clustering algorithm, gradually merging grid points with similar features to form several independent clusters. The system defines the set of grid points corresponding to each cluster as a spatial region candidate unit, and all spatial region candidate units together constitute a spatial region candidate set. The system verifies the internal feature similarity of each spatial region candidate unit to ensure that the grid points within each candidate unit maintain a high degree of similarity in climate background, terrain, and real-time meteorological characteristics, thus meeting the conditions for being an independent analysis unit.

[0063] S3: Based on wind field data in the spatiotemporal data cube, the transport trajectory of the catalyst in the target cloud system is simulated using the Lagrange particle diffusion model. The region with the highest overlap with the simulated diffusion path is selected from the candidate set of spatial regions to generate the optimal target area.

[0064] Specifically, the system constructs a Lagrange particle diffusion model. The core principle of this model is to simulate the transport and diffusion process of the catalyst in the target cloud system by tracking the trajectory of individual catalyst particles in the wind field. The system sets relevant initial and operational parameters for the Lagrange particle diffusion model. The initial release position of the catalyst particles is the cloud base height of the work area, and the initial release time is consistent with the start time of the operation. The initial particle size and initial diffusion rate are set according to the catalyst characteristics and operational requirements. The dynamic field input of the model is wind field data in a spatiotemporal data cube. The temporal resolution of the wind field data is consistent with the time step of the spatiotemporal data cube to ensure the accuracy of the model simulation. The system also sets the particle settling rate, diffusion coefficient, and survival time to ensure that the simulated trajectory can fully cover the key periods before and after the operation, comprehensively reflecting the catalyst transport and diffusion process. After completing the model construction and parameter settings, the system uses a numerical solution method based on the Lagrange particle diffusion model to solve the particle motion equations, obtaining the continuous transport trajectory of each catalyst particle during the key periods before and after the operation. The trajectory output is a latitude-longitude time series, which can clearly reflect the spatial location of the catalyst at different time points.

[0065] Furthermore, using the simulated catalyst transport trajectory as a benchmark, the system calculates the overlap degree between each candidate unit in the spatial region candidate set and the trajectory. The overlap degree calculation is based on the spatial intersection range between the trajectory and the candidate unit. The overlap degree index is obtained by calculating the proportion of the intersection range to the total trajectory range, which quantifies the degree of matching between the candidate unit and the catalyst transport trajectory. The system sorts all candidate units by overlap degree and selects the candidate unit with the highest overlap degree as the optimal target region. The optimal target region accurately covers the main transport and diffusion regions of the catalyst.

[0066] S4: In the candidate set of spatial regions, calculate and select another region with the highest comprehensive similarity to the optimal target region in terms of climate background and real-time meteorological characteristics, and generate the optimal comparison region.

[0067] Specifically, the system constructs a comprehensive similarity calculation model. This model quantifies the similarity between the optimal target area and each candidate unit in terms of climate background features and real-time meteorological features. The model calculates the comprehensive similarity using a weighted summation method. By setting a reasonable weight allocation, the influence of climate background features and real-time meteorological features on the comprehensive similarity is balanced, ensuring that the comprehensive similarity can comprehensively and objectively reflect the overall similarity between the two. Based on the constructed comprehensive similarity calculation model, the system calculates the similarity of climate background features and real-time meteorological features between the optimal target area and each candidate unit. The similarity of climate background features is obtained by comparing the climate background feature vectors of the two units, using a vector similarity calculation method to quantify the consistency between the two units under long-term climate background. The similarity of real-time meteorological features is obtained by comparing the real-time meteorological feature vectors of the two units, using a reasonable similarity calculation method to quantify the consistency between the two units under real-time meteorological conditions such as cloud systems and wind fields during the operation period.

[0068] Furthermore, the system substitutes the calculated similarity of climate background features and real-time meteorological features into the comprehensive similarity calculation model to obtain the comprehensive similarity between each candidate unit and the optimal target area. The system sorts the comprehensive similarity of all candidate units and selects the candidate unit with the highest comprehensive similarity as the optimal comparison area. The system verifies the characteristics of the optimal comparison area to ensure that the optimal comparison area and the optimal target area maintain a high degree of consistency in climate background and real-time meteorological features, and are not affected by catalyst transport and diffusion. This ensures that the optimal comparison area can truly reflect the background variation law of natural weather processes, providing a reliable comparison benchmark for subsequent calculation of net catalytic effect, and guaranteeing the scientificity and accuracy of the operational effect evaluation results.

[0069] S5: Based on the optimal target area and the optimal comparison area, extract the precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube, calculate the index time series of the optimal target area and the optimal comparison area respectively, and perform background value subtraction to generate the catalytic effect difference time series that characterizes the net effect of the catalytic operation.

[0070] Specifically, the system first extracts indicator sequences for the optimal target area and optimal comparison area from the spatiotemporal data cube. These extracted indicator sequences include precipitation and cloud microphysical indicators, both of which comprehensively reflect changes in precipitation and cloud physical processes in the target and comparison areas before and after the operation. The time span of the indicator sequences covers the critical periods from before to after the operation, and the time step is consistent with the time step of the spatiotemporal data cube, ensuring that the indicator sequences can completely capture the changes in meteorological parameters before and after the operation. After extracting the indicator sequences, the system performs background value subtraction on the indicator sequences for both types of areas to eliminate the impact of background fluctuations in natural weather processes on indicator changes. The system selects a continuous period before the operation as the baseline period and calculates the average value of the indicator sequences within the baseline period as the background value. This background value characterizes the normal level of indicator variation under natural weather processes.

[0071] Furthermore, the system subtracts the corresponding background value from the measured index value at each time point to obtain the index sequence after background value deduction. This sequence reflects the degree of deviation of the index value from the natural background, initially demonstrating the potential effects of catalytic operations. After background value deduction, the system calculates the difference between the index sequences of the optimal target area and the optimal comparison area after background value deduction, generating a catalytic effect difference time series. The difference is calculated by subtracting the index values ​​at corresponding time points. The difference at each time point characterizes the net effect of the catalytic operation at that moment, and the entire time series fully reflects the change process of the net effect of the catalytic operation over time. This catalytic effect difference time series eliminates the influence of background fluctuations in natural weather processes and excludes interference from regional differences, accurately characterizing the net effect of catalytic operations on precipitation and cloud microphysical processes.

[0072] S6: Perform integrated calculations on the time series of catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate an evaluation conclusion on the operation effect.

[0073] Specifically, the system employs a time-weighted integral method to integrate and calculate the time series of catalytic effect differences, obtaining an integrated effect value. The time-weighted integral method sets time weights based on the contribution of different time points to the catalytic effect, with the weight of the post-operation period higher than that of the pre-operation period. This ensures that the contribution of the post-operation catalytic effect is fully reflected, and the integrated effect value comprehensively reflects the total net effect generated by the catalytic operation throughout the entire critical period, achieving a quantitative summary of the catalytic effect.

[0074] After calculating the integrated effect value, the system determines a significance threshold. This threshold is used to determine whether the integrated effect value exceeds the normal fluctuation range of natural weather processes, thereby determining whether the catalytic operation has produced a significant effect. The significance threshold is generated based on historical operation data. The system first collects the difference data of catalytic effects from similar past artificial rain enhancement operations to construct a historical dataset. The system performs statistical analysis on the historical dataset, using a probability distribution fitting method to fit the historical data. Based on the fitting results, a significance threshold is determined. This threshold can characterize the extreme range of natural weather process fluctuations. If the integrated effect value exceeds this threshold, it indicates that the effect of the catalytic operation exceeds natural fluctuations. The system compares the calculated integrated effect value with the significance threshold and generates a graded evaluation conclusion of the operation effect based on the comparison results. Simultaneously, the system quantifies the uncertainty brought about by subjective selection during the evaluation process by calculating the dispersion of the comprehensive similarity between the optimal target area and each candidate unit. The uncertainty quantification result can reflect the reliability of the evaluation conclusion and provide a reference for the application of the evaluation conclusion.

[0075] In summary, the artificial rain enhancement operation effect evaluation method based on the index threshold method provided in this application can completely replace the subjective operation of manually delineating areas based on experience by constructing a standardized spatiotemporal data cube and an objective region candidate set, eliminating the systematic bias introduced by human selection from the data source. By integrating the physical simulation of the Lagrange particle diffusion model with multi-dimensional feature dynamic matching, it can automatically and adaptively determine the optimal target area and comparison area with clear physical meaning and comparable climatic background according to the specific weather conditions and geographical environment of each operation, effectively overcoming the shortcomings of traditional methods in terms of poor applicability under complex terrain and rapidly evolving weather systems. Based on this, the method can achieve precise stripping and quantification of the net effect of catalytic operations, and generate quantitative and reliable operation effect evaluation conclusions by objectively comparing with statistical significance thresholds. Ultimately, this automated process not only significantly improves the objectivity and accuracy of the evaluation results, but also inherently contains the ability to analyze the uncertainty of the evaluation conclusions, providing solid and reliable technical support for the scientific evaluation and decision-making of artificial rain enhancement operations.

[0076] In one embodiment, S1 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0077] S11: Perform gridded interpolation on the static geographic background data in the raw data containing the geographic background of the work area, real-time meteorological observations, and numerical forecasts. This converts the discrete geographic information data into a continuous background field with the same spatial resolution as the meteorological observation data, generating a standardized static background data layer.

[0078] Specifically, the system performs specialized processing on the static geographic background data in the original data. This static geographic background data refers to the fixed, unchanging data within the geographic background data of the work area, encompassing topographical, geomorphological, and administrative division data of the work area and its surroundings. This type of data often exists in discrete form and cannot be directly matched and fused with continuous meteorological observation data. Therefore, the system needs to achieve data format conversion and resolution unification through gridded interpolation. The system extracts the static geographic background data from the original data, preprocesses the extracted discrete geographic information data, and removes invalid data to ensure data integrity and accuracy. The system determines the core parameters for gridded interpolation, ensuring that the spatial resolution of the interpolation grid strictly matches the spatial resolution of the meteorological observation data, guaranteeing precise spatial alignment between the processed static geographic background data and the meteorological observation data.

[0079] Preferably, the system can employ an adapted gridded interpolation method to interpolate discrete geographic information data. The interpolation algorithm fills each grid point with discretely distributed geographic information data, transforming discrete data into continuous data. During interpolation, the system ensures that the geographic information data at each grid point accurately reflects the actual geographic characteristics of the grid area, avoiding the impact of interpolation bias on subsequent data fusion and analysis. After interpolation, the system generates a continuous background field covering the entire operational area and its surrounding region. Each grid point contains complete static geographic background information with a spatial resolution consistent with meteorological observation data.

[0080] S12: Perform spatiotemporal alignment and quality control processing on the real-time meteorological observation and numerical forecast data in the raw data, remove outliers, and uniformly interpolate data from different sources and at different times to the same spatiotemporal grid point to generate a standardized dynamic meteorological data layer.

[0081] Specifically, the system performs spatiotemporal alignment and quality control processing on real-time meteorological observation data and numerical forecast data in the raw data. These two types of data come from different sources, have different formats, and occur at different times, and may contain outliers. Standardization processing is required to meet the requirements of subsequent data fusion and analysis. The system categorizes and extracts real-time meteorological observation data and numerical forecast data, clarifying the source attributes and data characteristics of the two types of data. Real-time meteorological observation data comes from various meteorological observation equipment and networks, while numerical forecast data comes from various numerical weather prediction models. Both types of data contain multiple meteorological-related parameters.

[0082] Furthermore, the system first performs quality control, verifying both types of data one by one according to preset quality control rules, identifying and removing outlier data. Outlier identification is based on the reasonableness range of the data and its variation patterns. For any missing data resulting from outlier removal, the system uses an appropriate interpolation method to fill in the gaps, ensuring data continuity. After quality control, the system performs spatiotemporal alignment. During temporal alignment, the system unifies the two types of data from different time points to the same time standard, adjusting the temporal resolution of the data through a time interpolation algorithm to ensure consistent time steps for both types of data. During spatial alignment, the system selects a spatial grid consistent with the static background data layer as a unified reference, and uses a spatial interpolation algorithm to uniformly interpolate the two types of data from different sources and with different spatial resolutions to each grid point of this spatiotemporal grid.

[0083] S13: Perform multi-dimensional fusion processing on the static background data layer and the dynamic meteorological data layer. According to the grid coordinates and time series, the two types of data are superimposed and integrated into a multi-dimensional array containing geography, climate and real-time meteorological status, generating a standardized spatiotemporal data cube for the entire domain.

[0084] Specifically, based on the generated standardized static background data layer and standardized dynamic meteorological data layer, the system performs multi-dimensional fusion processing to achieve seamless integration of the two types of data. It integrates geospatial information and dynamic meteorological information to construct a standardized spatiotemporal data cube covering the entire domain, providing a unified data foundation for subsequent evaluation steps. The system first performs a data consistency check on the two types of data, verifying whether their spatial grid references, data formats, and data precision are consistent to ensure fusion is feasible. If inconsistencies exist, the system performs a secondary standardization adjustment until the fusion requirements are met.

[0085] After successful verification, the system performs an overlay and integration operation on the two types of data according to grid coordinates and time series. Spatially, the system uses a unified latitude and longitude grid as a benchmark, associating and matching the geospatial information of each grid point in the static background data layer with various meteorological parameters of the corresponding grid points in the dynamic meteorological data layer, ensuring that each grid point simultaneously contains both geospatial and meteorological characteristics. Temporally, the system overlays the time series data of the dynamic meteorological data layer with the static background data layer according to a unified time step and time series. The static background data layer serves as the base data throughout the entire time series, while the dynamic meteorological data layer is overlaid onto the grid points of each time node according to temporal changes. The system then converts the overlaid and integrated data into a multidimensional array format, which includes spatial, temporal, and indicator dimensions, forming a complete dataset containing geographic, climatic, and real-time meteorological conditions.

[0086] In one embodiment, S2 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0087] S21: Perform feature dimension traversal and calculation processing on the spatiotemporal data cube, extract feature values ​​of multiple dimensions such as elevation, historical precipitation and climate values, surface type, and reflectivity, brightness temperature, and wind field at the start of the operation for each grid point, and generate a comprehensive feature vector for each grid point.

[0088] Specifically, for each grid point, the system extracts different types of feature values. Elevation and surface type feature values ​​are extracted from the static background data layer of the spatiotemporal data cube, directly corresponding to the inherent geographical attributes of the grid point. Historical precipitation and climate values ​​are extracted from the climate background data of the spatiotemporal data cube, obtained through statistical calculations of historical precipitation data in the grid point's region, reflecting the grid point's long-term precipitation background characteristics. Reflectivity, brightness temperature, wind field, and other feature values ​​at the start of the operation are extracted from the dynamic meteorological data layer of the spatiotemporal data cube, corresponding to the grid point's real-time meteorological status at the initial moment of the operation. The system integrates all extracted feature values ​​in a preset order to generate a unique comprehensive feature vector for each grid point. Each comprehensive feature vector covers three core dimensions: geography, climate, and real-time meteorology, ensuring a comprehensive representation of the grid point's overall characteristics.

[0089] S22: Perform unsupervised classification based on density clustering on the comprehensive feature vectors of all grid points. Automatically identify and merge grid points with similar features according to the distribution density of feature vectors in the multivariate space, and generate multiple spatial grid point clusters with high internal homogeneity.

[0090] Specifically, unsupervised classification does not require pre-setting classification categories and quantities; it relies entirely on the inherent characteristics and distribution patterns of the data to achieve automatic classification. This avoids subjective bias caused by manually pre-setting categories and ensures the objectivity of the classification results. The system first standardizes the comprehensive feature vectors of all grid points, eliminating dimensional differences between different feature dimensions. This ensures that all types of features have equal weight in clustering calculations, avoiding clustering bias caused by dimensional differences.

[0091] Preferably, the system employs a density clustering algorithm, using standardized composite feature vectors as input data. The core logic of the algorithm is grid merging based on the distribution density of feature vectors in the multivariate space. The system identifies high-density feature vector clusters by calculating the distribution density of other feature vectors surrounding each feature vector, and groups grid points corresponding to feature vectors within the same high-density region with high feature similarity into one class. During the clustering process, the system automatically adjusts the clustering parameters to ensure that the feature vectors corresponding to each class of grid points are concentrated in the multivariate space, ultimately generating multiple spatial grid point clusters. The composite features of grid points within each spatial grid point cluster are highly homogeneous, while the feature differences between different grid point clusters are significant.

[0092] S23: Perform geographic connectivity and meteorological rationality verification on the spatial grid clusters, eliminate discrete noise points, and encapsulate the spatial range vectorization of the grid clusters that meet the verification to generate a candidate set of spatial regions.

[0093] Specifically, the system performs geographic connectivity verification by analyzing the spatial coordinate correlation of all grid points within each spatial grid cluster. This checks whether the grid points within a cluster possess continuous spatial distribution characteristics, eliminating discrete noise points caused by clustering errors and disconnected, scattered grid points, ensuring the continuity of the spatial distribution of each grid cluster. Further, the system performs meteorological rationality verification by comparing the meteorological characteristic values ​​of all grid points within each cluster, checking the consistency of meteorological characteristics within the cluster, and eliminating grid clusters that do not conform to natural meteorological change patterns or have obvious contradictions in their meteorological characteristics. After passing the verification, the system performs spatial range vectorization encapsulation on the qualified spatial grid clusters. By extracting the coordinates of the boundary grid points of each cluster, the system constructs the spatial boundary contour of the cluster, converting the discrete spatial grid clusters into vector regions with a defined spatial range. After all qualified grid clusters have been vectorized, the system integrates them to generate a candidate set of spatial regions. Each vector region serves as an independent potential analysis unit, providing sufficient candidates for subsequent selection of optimal target areas and comparison areas.

[0094] In one embodiment, S3 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0095] S31: Perform vertical integration and smoothing on the multi-level wind field data in the spatiotemporal data cube, calculate the guiding airflow field representing the average catalyst delivery path, and generate the driving wind field for particle diffusion simulation.

[0096] Specifically, the system calls upon multi-level wind field data from the spatiotemporal data cube. This data originates from the dynamic meteorological data layer and covers wind field-related parameters at different vertical heights. Vertical integration and smoothing are required to eliminate data fluctuations and generate a stable driving wind field, providing dynamic support for subsequent particle diffusion simulations. The system first extracts wind field data from all vertical levels within the spatiotemporal data cube, performs consistency checks on the extracted multi-level wind field data, and removes invalid data to ensure data integrity. Subsequently, the system performs vertical integration processing, with the vertical integration calculation following the formula:

[0097]

[0098] in, This is the vertical integral wind field vector. This represents the highest vertical height of the wind field data. This represents the minimum vertical height for wind field data. Let z be the wind field vector at height z. The system integrates the wind field data at different vertical levels according to this formula, integrating the wind field characteristics at each level to obtain the integral result representing the average wind field state in the vertical direction. After completing the vertical integration, the system performs smoothing processing on the integration result, eliminating random and abnormal fluctuations in the wind field data through a smoothing algorithm, reducing the impact of data noise on subsequent simulations. Based on the smoothed vertical integration result, the system calculates the steering airflow field, standardizes the steering airflow field, unifies the data format and accuracy, and finally generates the driving wind field for particle diffusion simulation, ensuring the accuracy and reliability of the particle diffusion simulation.

[0099] S32: Based on the driving wind field, Lagrange particle reverse trajectory simulation is performed. A large number of virtual particles are released from the operation point, and the three-dimensional backward trajectory and spatial residence probability of the virtual particles within a certain period of time are calculated based on the wind field data to generate the potential influence range probability field of the catalyst.

[0100] Specifically, the system uses the generated driving wind field as its power basis to simulate the backward trajectory of Lagrange particles. The core objective is to determine the potential influence range of the catalyst through virtual particle trajectory simulation, providing a basis for selecting the optimal target area. The system first determines the relevant simulation parameters, clearly defining the release location of the virtual particles as the operating point, and ensuring the release quantity meets the simulation accuracy requirements. Subsequently, the system releases a large number of virtual particles from the operating point. Based on the Lagrange particle diffusion model, the system calculates the three-dimensional backward trajectory of each virtual particle using the driving wind field data. The trajectory calculation follows the particle motion equations:

[0101]

[0102] in, Let be the spatial position vector of the virtual particle at time t. To drive the wind field vector, Let be the particle's random diffusion velocity vector. The system performs statistical analysis on the three-dimensional backward trajectories of all virtual particles, calculating the spatial dwell probability at each spatial grid point. The dwell probability calculation follows the formula:

[0103]

[0104] in, Let be the dwell probability of the grid point (x,y), and N be the total number of virtual particles. Let be the dwell time of the i-th virtual particle at grid point (x, y). The system integrates the dwell probabilities of all grid points to generate a probability field for the potential influence range of the catalyst.

[0105] S33: Perform spatial overlay and overlap analysis on the probability field of potential influence range and the candidate set of spatial regions, calculate the spatial fit between each candidate region and the high value region of the probability field of potential influence range, and determine the candidate region with the highest spatial fit as the optimal target region.

[0106] Specifically, the system calls the generated potential influence range probability field and the generated spatial region candidate set, performs spatial overlay and overlap analysis, and selects the optimal target area by calculating the spatial fit. The system first performs spatial overlay processing on the potential influence range probability field and the spatial region candidate set to ensure precise alignment on the spatial grid reference, so that each candidate region can accurately match the corresponding grid point in the probability field. Then, the system identifies high-value regions in the potential influence range probability field, which consist of all grid points in the probability field with probability values ​​higher than a preset probability threshold. The system calculates the spatial fit between each candidate region and the high-value region. The formula for calculating the spatial fit is:

[0107]

[0108] in, Candidate region Space adaptability, For grid points The probability value of the catalyst at that location. This refers to the high-probability region comprised of all grid points in the probability field whose probability values ​​exceed a preset probability threshold. Spatial weights related to the distance from the grid point to the work point and the terrain uplift effect, Candidate region The total number of grid points included. The system sorts the spatial fit of all candidate regions, selects the candidate region with the highest spatial fit, and determines it as the optimal target region, ensuring that the optimal target region can accurately cover the core area of ​​the catalyst's potential impact.

[0109] In one embodiment, S4 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0110] S41: After excluding the optimal target area from the spatial region candidate set, representative feature extraction processing is performed on the remaining candidate areas. The static feature mean vector and the change sequence of dynamic meteorological features before and after the operation are calculated for each area to generate a standardized feature profile of each candidate area.

[0111] Specifically, the system calls the spatial region candidate set and performs an exclusion operation against the determined optimal target region, removing the optimal target region from the spatial region candidate set and retaining all remaining candidate regions as comparison areas for screening. The system then performs representative feature extraction processing on each remaining candidate region, relying on the static background data layer and dynamic meteorological data layer in the spatiotemporal data cube. The system calculates the static feature mean vector for each candidate region, which is obtained using the following formula:

[0112]

[0113] in, Let be the static feature mean vector of the i-th candidate region, and M be the total number of grid points contained in the candidate region. This represents the static feature vector of the k-th grid point within the candidate region. Simultaneously, the system extracts dynamic meteorological feature data for each candidate region before and after the operation, generating a dynamic meteorological feature change sequence to characterize the evolution of meteorological parameters before and after the operation. The system standardizes the static feature mean vector and the dynamic meteorological feature change sequence to eliminate the influence of dimensions, integrating the two to form a standardized feature profile for each candidate region. This profile comprehensively covers the static attributes and dynamic meteorological change characteristics of the candidate region.

[0114] S42: Perform multi-dimensional similarity calculations between the standardized feature profiles of each candidate region and the corresponding features of the optimal target region, comprehensively evaluate the degree of matching between the two in terms of climate background, topographic attributes and the temporal evolution of key meteorological indicators, and generate a comprehensive similarity score for each candidate region relative to the target region.

[0115] Specifically, the system calls upon the standardized feature profiles of each candidate region and the corresponding features of the optimal target region, performing multi-dimensional similarity calculations to evaluate the degree of matching from three dimensions: climate background, topographic attributes, and the temporal evolution of key meteorological indicators. The similarity calculation strictly follows a preset formula, where the formula for calculating the comprehensive similarity score is:

[0116]

[0117] in, The score represents the combined similarity between candidate region i and the optimal target region T. and These are the static background feature vectors of the target region T and the candidate region i, respectively. Let be the variance of the distances between the static feature vectors of all candidate regions. and These are time series of key dynamic meteorological indicators for the target area T and candidate area i within a preset time period before the operation. For all candidate regions and target region T The maximum distance, and To adjust the weighting coefficients of static and dynamic similarity contributions, their sum is 1. The system integrates static and dynamic similarity using this formula to generate a comprehensive similarity score for each candidate region relative to the optimal target region.

[0118] S43: Compare, sort, and filter the comprehensive similarity scores of all candidate regions, select the candidate region with the highest score that exceeds the preset similarity threshold, and generate the optimal comparison region.

[0119] Specifically, the system collects the comprehensive similarity scores of all remaining candidate regions, compares, sorts, and filters them to select the candidate region that best matches the characteristics of the optimal target region as the optimal comparison region, ensuring that the comparison region can truly reflect the background response of natural weather processes. The system first sorts all comprehensive similarity scores, arranging the candidate regions in descending order of score to clarify the matching priority of each candidate region with the optimal target region. After sorting, the system calls a preset similarity threshold, which is used to determine whether the matching degree between the candidate region and the optimal target region meets the requirements. Only candidate regions with scores exceeding this threshold are eligible to be considered as optimal comparison regions. The system filters the sorted candidate regions one by one, selecting the candidate region with the highest score that exceeds the preset similarity threshold. This candidate region undergoes feature verification to confirm that it has not been affected by catalyst transport and diffusion and that it highly matches the optimal target region in terms of climate background, topographic attributes, and dynamic meteorological changes. After successful verification, the system determines this candidate region as the optimal comparison region, completing the generation of the optimal comparison region.

[0120] In one embodiment, S5 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0121] S51: Based on the spatial boundaries of the optimal target area and the optimal comparison area, spatial mask extraction processing is performed on the spatiotemporal data cube. Precipitation rate, radar combined reflectivity and other index data of all grid points in the two areas during the baseline period before operation and the evaluation period after operation are extracted respectively to generate the original spatiotemporal dataset of the optimal target area and the optimal comparison area.

[0122] Specifically, the system calls upon the spatial boundary data of the determined optimal target area and optimal comparison area, as well as the globally standardized spatiotemporal data cube, to perform spatial mask extraction processing, extracting relevant indicator data from the two areas to provide basic data for subsequent background subtraction and difference calculation. The system first converts the spatial boundaries of the optimal target area and optimal comparison area into a mask template that matches the spatiotemporal data cube. The generation of the mask template follows the formula:

[0123]

[0124] in, For grid points The mask value is used, where A represents the spatial boundary of the target or comparison area. A mask value of 1 indicates that the grid point belongs to the area, while 0 indicates that it does not. The system spatially matches the mask template with the spatiotemporal data cube, extracting indicator data from all grid points in both areas during the baseline period before the operation and the evaluation period after the operation. The selected indicators include core meteorological indicators such as precipitation rate and radar combined reflectivity. The system performs integrity verification on the extracted indicator data, removes invalid data, and categorizes and organizes the verified indicator data by region, generating original spatiotemporal datasets for the optimal target area and the optimal comparison area, ensuring that the datasets can fully represent the indicator change characteristics of the two areas during key periods.

[0125] S52: Perform time series aggregation on the original spatiotemporal dataset of indicators, calculate the average indicator values ​​of the optimal target area and the optimal comparison area at each evaluation time, and remove the average state value of each time period before the operation baseline period from the value of each time period to generate the indicator increment sequence after background subtraction for the optimal target area and the optimal comparison area.

[0126] Specifically, the system calls upon the original spatiotemporal datasets of the optimal target area and the optimal comparison area, performs time-series aggregation processing, and integrates the grid point index data within the region. The system aggregates and calculates the original spatiotemporal datasets of each region according to the evaluation time, and the average index value is calculated using the following formula:

[0127]

[0128] in, Let be the average index value of region A at the t-th evaluation time. This represents the total number of grid points contained in region A. Let be the index value of the k-th grid point in region A at time t. The system uses this formula to calculate the average index value for each evaluation time in both regions, generating an average index time series. Subsequently, the system calculates the average state value of both regions during the baseline period before the operation, subtracting the baseline average state value of the corresponding region from the average index value for each evaluation time to obtain the index increment series after background subtraction. The system standardizes the index increment series to eliminate the influence of dimensions, generating index increment series for the optimal target area and the optimal comparison area, respectively. This series reflects the degree of deviation of the index from the natural background.

[0129] S53: Perform time-by-time difference processing on the index increment sequence, calculate the difference between the index increment of the optimal target area and the index increment of the optimal comparison area at the same time, and generate a time series of catalytic effect difference that characterizes the net effect of catalytic operation.

[0130] Specifically, the system calls the indicator increment sequences after background subtraction from the optimal target area and the optimal comparison area, performs hourly differencing to eliminate the impact of background fluctuations in natural weather processes, and extracts the net effect generated by the catalytic operation. The system first precisely aligns the indicator increment sequences of the two regions along the time axis to ensure that increment values ​​at the same time point correspond and match, avoiding calculation errors caused by time misalignment. Further, the system performs hourly differencing calculations on the aligned indicator increment sequences, with the difference calculation following the formula:

[0131]

[0132] in, Let be the difference in catalytic effect at time t. This represents the incremental value of the indicator after background subtraction for the optimal target area at time t. This represents the incremental value of the indicator after background subtraction in the optimal comparison area at time t. The system calculates the difference in catalytic effect at each time using this formula, and integrates the differences from all time periods in chronological order to generate a time series of catalytic effect differences. This series can accurately characterize the net impact of catalytic operations on relevant meteorological indicators, eliminating the interference of natural background fluctuations.

[0133] In one embodiment, step S6 of the artificial rain enhancement operation effect evaluation method based on the index threshold method provided by the present invention specifically includes the following steps:

[0134] S61: Perform time-weighted integral processing based on physical processes on the time series of catalytic effect difference. Assign different contribution weights to the values ​​at different times in the series according to the typical time period characteristics of the occurrence and development of catalysis. Then, accumulate the weighted values ​​to generate the integrated effect value.

[0135] Specifically, the system calls the generated time series of catalytic effect differences, performs time-weighted integral processing based on the physical process, integrates the net catalytic effects at different times, and generates an integrated effect value that comprehensively reflects the overall effect of the operation. The system first analyzes the typical time period characteristics of the catalytic effect's occurrence and development, identifying the onset, peak, and decay periods of the catalytic effect. Based on these characteristics, it assigns a corresponding contribution weight to each time point in the time series, with the weight of the peak period being higher than that of the onset and decay periods, ensuring that the weight allocation conforms to the laws of the catalytic physical process. The weighted integral calculation follows the formula:

[0136]

[0137] in, This is the integrated effect value. Let be the contribution weight at time t, and let the sum of the weights at all times be 1. Let be the difference in catalytic effect at time t, and n be the total number of times in the time series of catalytic effect difference. The system uses this formula to accumulate the weighted difference for each time period, and then performs standardized verification on the calculation results to ensure accuracy and reliability, generating an integrated effect value that quantitatively reflects the overall net effect of the catalytic operation.

[0138] S62: Perform statistical comparative analysis on the integrated effect value and the preset significance threshold to determine whether the integrated effect value exceeds the critical level defined by the preset significance threshold, and generate a qualitative judgment result on whether the operation effect is significant.

[0139] Specifically, the system calls the generated integrated effect value and a preset significance threshold, performing statistical comparative analysis. The core purpose is to determine whether the net effect of the catalytic operation exceeds the normal fluctuation range of natural weather processes. The significance threshold is generated based on historical data from similar operations, following the formula:

[0140]

[0141] in, The significance threshold, This represents the mean of the ensemble effect values ​​of similar historical tasks. denoted as , where is the standard deviation of the historical integrated effect values, and k is the coefficient corresponding to the statistical confidence level. The system statistically compares the integrated effect value with the significance threshold, analyzes the numerical difference and its statistical significance, and determines whether the integrated effect value exceeds the critical level defined by the significance threshold. If the integrated effect value exceeds the critical level, the system generates a qualitative judgment result indicating that the operation effect is significant; if it does not exceed the critical level, the system generates a qualitative judgment result indicating that the operation effect is not significant. The qualitative judgment result accurately reflects the statistical significance of the catalytic operation effect.

[0142] S63: Based on the qualitative judgment results, organize and output the conclusions, integrate the qualitative judgment results and the integrated effect value into a structured evaluation statement, and generate the operation effect evaluation conclusion. The operation effect evaluation conclusion is used to indicate whether the artificial rain enhancement operation has produced a statistically significant catalytic effect.

[0143] Specifically, the system calls upon the generated qualitative judgment results and the generated integrated effect value, organizes and outputs the conclusions, and its core purpose is to generate a structured and standardized evaluation conclusion of the operation's effectiveness, providing a clear basis for judging the operation's effectiveness. The system first performs a consistency check on the qualitative judgment results and the integrated effect value to ensure logical consistency and avoid contradictions between data and judgments. After the check passes, the system integrates the two into a structured evaluation statement. The evaluation statement clearly includes the specific value of the integrated effect value, the qualitative judgment result, and a brief explanation of the judgment basis, namely the comparison result between the integrated effect value and the significance threshold. The system standardizes the structured evaluation statement, unifying the expression format and specifications to ensure that the evaluation conclusion is clear, rigorous, and interpretable. After processing, the system generates an operation effectiveness evaluation conclusion, which directly indicates whether the artificial rain enhancement operation has produced a statistically significant catalytic effect.

[0144] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0145] Based on the same inventive concept, this application also provides a system for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method, which is used to implement the aforementioned method for evaluating the effectiveness of artificial rain enhancement operations. The solution provided by this system is similar to the implementation described in the above method. Therefore, the specific limitations of one or more embodiments of the system for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method provided below can be found in the limitations of the method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method described above, and will not be repeated here.

[0146] Preferably, such as Figure 2 As shown, this invention provides an artificial rain enhancement operation effect evaluation system 700 based on the index threshold method. This system is configured with the following modules:

[0147] The spatiotemporal data standardization and fusion module 710 is used to standardize and fuse raw data containing geographical background of the work area, real-time meteorological observations and numerical forecasts, align the raw data with spatiotemporal references and unify the format, and generate a standardized spatiotemporal data cube covering the entire domain.

[0148] The multidimensional feature region candidate generation module 720 is used to extract multidimensional features from the spatiotemporal data cube, automatically classify the grid points based on the climate background, terrain and real-time meteorological features, and generate multiple spatial region candidate sets with similar internal features that can be used as potential analysis units.

[0149] The optimal target area screening module 730 is used to simulate the transport trajectory of the catalyst in the target cloud system based on the wind field data in the spatiotemporal data cube and the Lagrange particle diffusion model, and to select the region with the highest overlap with the simulated diffusion path from the spatial region candidate set to generate the optimal target area.

[0150] The optimal comparison area selection module 740 is used to calculate and select the other region with the highest comprehensive similarity to the optimal target region in terms of climate background and real-time meteorological characteristics from the spatial region candidate set, and generate the optimal comparison area.

[0151] The catalytic effect difference sequence generation module 750 is used to extract precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube based on the optimal target area and the optimal comparison area. It calculates the index time series of the optimal target area and the optimal comparison area respectively and performs background value subtraction to generate a catalytic effect difference time series characterizing the net effect of the catalytic operation.

[0152] The task effectiveness evaluation module 760 is used to perform integrated calculations on the time series of catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate a task effectiveness evaluation conclusion.

[0153] Preferably, the spatiotemporal data standardization and fusion module 710 provided in this application is configured with the following units:

[0154] The static geographic background grid interpolation unit is used to perform grid interpolation processing on the static geographic background data in the raw data containing the geographic background of the work area, real-time meteorological observations and numerical forecasts. It converts discrete geographic information data into a global continuous background field with the same spatial resolution as the meteorological observation data, and generates a standardized static background data layer.

[0155] The dynamic meteorological data spatiotemporal calibration unit is used to perform spatiotemporal alignment and quality control processing on real-time meteorological observation and numerical forecast data in the raw data, remove outliers, and uniformly interpolate data from different sources and at different times to the same spatiotemporal grid point to generate a standardized dynamic meteorological data layer.

[0156] The spatiotemporal data cube fusion generation unit is used to perform multidimensional fusion processing on the static background data layer and the dynamic meteorological data layer. According to the grid coordinates and time series, the two types of data are superimposed and integrated into a multidimensional array containing geography, climate and real-time meteorological status, generating a standardized spatiotemporal data cube for the entire domain.

[0157] Preferably, the multi-dimensional feature region candidate generation module 720 provided in this application is configured with the following units:

[0158] The grid-based comprehensive feature extraction unit is used to traverse and calculate the feature dimensions of the spatiotemporal data cube. For each grid point, it extracts feature values ​​of multiple dimensions such as elevation, historical precipitation and climate values, land surface type, and reflectivity, brightness temperature, and wind field at the start of the operation, and generates a comprehensive feature vector for each grid point.

[0159] The grid clustering grouping unit is used to perform unsupervised classification processing based on density clustering on the comprehensive feature vector of all grid points. It automatically identifies and merges grid points with similar features according to the distribution density of feature vectors in the multidimensional space, generating multiple spatial grid clusters with high internal homogeneity.

[0160] The regional candidate set generation unit is used to perform geographic connectivity and meteorological rationality verification on spatial grid clusters, eliminate discrete noise points, and perform spatial range vectorization encapsulation on grid clusters that meet the verification to generate a spatial regional candidate set.

[0161] Preferably, the optimal target area screening module 730 provided in this application is configured with the following units:

[0162] The guiding airflow field generation unit is used to perform vertical integration and smoothing on multi-level wind field data in the spatiotemporal data cube, calculate the guiding airflow field representing the average catalyst delivery path, and generate the driving wind field for particle diffusion simulation.

[0163] The potential impact range probability field generation unit is used to perform Lagrange particle reverse trajectory simulation based on the driving wind field. It releases a large number of virtual particles from the operation point and calculates the three-dimensional backward trajectory and spatial residence probability of the virtual particles within a certain period of time based on the wind field data, thereby generating the potential impact range probability field of the catalyst.

[0164] The optimal target area determination unit is used to perform spatial overlay and overlap analysis on the probability field of potential influence range and the set of candidate spatial regions, calculate the spatial fit between each candidate region and the high-value region of the probability field of potential influence range, and determine the candidate region with the highest spatial fit as the optimal target area.

[0165] Preferably, the optimal comparison area screening module 740 provided in this application is configured with the following units:

[0166] The candidate region feature extraction unit is used to extract representative features from the remaining candidate regions after excluding the optimal target region from the spatial region candidate set. It calculates the static feature mean vector and the change sequence of dynamic meteorological features before and after the operation for each region, and generates a standardized feature profile for each candidate region.

[0167] The regional similarity calculation unit is used to perform multi-dimensional similarity calculations on the standardized feature profiles of each candidate region and the corresponding features of the optimal target region. It comprehensively evaluates the degree of matching between the two in terms of climate background, topographic attributes and the temporal evolution of key meteorological indicators, and generates a comprehensive similarity score for each candidate region relative to the target region.

[0168] The optimal comparison area filtering unit is used to compare, sort, and filter the comprehensive similarity scores of all candidate areas, and select the candidate area with the highest score that exceeds the preset similarity threshold to generate the optimal comparison area.

[0169] Preferably, the catalytic effect difference sequence generation module 750 provided in this application is configured with the following units:

[0170] The spatiotemporal data mask extraction unit is used to perform spatial mask extraction processing on the spatiotemporal data cube based on the spatial boundaries of the optimal target area and the optimal comparison area. It extracts the precipitation rate, radar combined reflectivity and other index data of all grid points in the two areas during the baseline period before the operation and the evaluation period after the operation, respectively, and generates the original spatiotemporal dataset of the index of the optimal target area and the optimal comparison area.

[0171] The indicator data aggregation and deduction unit is used to perform time series aggregation on the original indicator spatiotemporal dataset, calculate the average indicator values ​​of the optimal target area and the optimal comparison area at each evaluation time, and remove the average state value of each time period before the operation baseline period from the value of each time period to generate the indicator increment sequence after background deduction of the optimal target area and the optimal comparison area.

[0172] The catalytic effect difference calculation unit is used to perform time-by-time difference processing on the index increment sequence, calculate the difference between the index increment of the optimal target area and the index increment of the optimal comparison area at the same time, and generate a time series of catalytic effect difference that characterizes the net effect of catalytic operation.

[0173] Preferably, the job performance evaluation module 760 provided in this application is configured with the following units:

[0174] The integrated effect value weighted calculation unit is used to perform time-weighted integration processing on the time series of catalytic effect difference based on physical processes. According to the typical time period characteristics of the occurrence and development of catalytic effect, different contribution weights are assigned to the values ​​at different times in the series, and the weighted values ​​are accumulated to generate the integrated effect value.

[0175] The effect significance judgment unit is used to perform statistical comparative analysis on the integrated effect value and the preset significance threshold, determine whether the integrated effect value exceeds the critical level defined by the preset significance threshold, and generate a qualitative judgment result on whether the operation effect is significant.

[0176] The operation effect conclusion generation unit is used to organize and output conclusions based on qualitative judgment results. It integrates qualitative judgment results and integrated effect values ​​into a structured evaluation statement to generate operation effect evaluation conclusions. The operation effect evaluation conclusions are used to indicate whether artificial rain enhancement operations have produced statistically significant catalytic effects.

[0177] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method.

[0178] In one embodiment, this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method.

[0179] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0180] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0181] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method, characterized in that, Includes the following steps: S1: Standardize and fuse the raw data containing the geographical background of the work area, real-time meteorological observations and numerical forecasts, align the raw data with spatiotemporal references and unify the format to generate a standardized spatiotemporal data cube covering the entire domain. S2: Perform multi-dimensional feature extraction on the spatiotemporal data cube, and automatically classify it according to the climate background, terrain and real-time meteorological features of the grid points to generate multiple spatial region candidate sets with similar internal features that can be used as potential analysis units. S3: Based on the wind field data in the spatiotemporal data cube, the transport trajectory of the catalyst in the target cloud system is simulated by the Lagrange particle diffusion model. The region with the highest overlap with the simulated diffusion path is selected from the candidate set of spatial regions to generate the optimal target area. S4: In the candidate set of spatial regions, calculate and select another region with the highest comprehensive similarity to the optimal target region in terms of climate background and real-time meteorological characteristics, and generate the optimal comparison region; S5: Based on the optimal target area and the optimal comparison area, extract the precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube, calculate the index time series of the optimal target area and the optimal comparison area respectively and perform background value subtraction to generate a time series of catalytic effect difference characterizing the net effect of catalytic operation. S6: Perform integrated calculations on the time series of the catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate an evaluation conclusion on the operation effect.

2. The method according to claim 1, characterized in that, S1 includes: S11: Perform gridded interpolation on the static geographic background data in the raw data containing the geographic background of the work area, real-time meteorological observations and numerical forecasts, and convert the discrete geographic information data into a global continuous background field with the same spatial resolution as the meteorological observation data, generating a standardized static background data layer. S12: Perform spatiotemporal alignment and quality control processing on the real-time meteorological observation and numerical forecast data in the original data, remove outliers, and uniformly interpolate data from different sources and at different times to the same spatiotemporal grid point to generate a standardized dynamic meteorological data layer. S13: Perform multi-dimensional fusion processing on the static background data layer and the dynamic meteorological data layer, and integrate the two types of data into a multi-dimensional array containing geography, climate and real-time meteorological status according to grid coordinates and time sequence, and generate a standardized spatiotemporal data cube.

3. The method according to claim 1, characterized in that, S2 includes: S21: Perform feature dimension traversal and calculation processing on the spatiotemporal data cube, extract feature values ​​of multiple dimensions such as elevation, historical precipitation climate value, land surface type, and reflectivity, brightness temperature, and wind field at the start of the operation for each grid point, and generate a comprehensive feature vector for each grid point. S22: Perform unsupervised classification processing based on density clustering on the comprehensive feature vector of all grid points, and automatically identify and merge grid points with similar features according to the distribution density of feature vectors in multidimensional space to generate multiple spatial grid point clusters with high internal homogeneity. S23: Perform geographic connectivity and meteorological rationality verification on the spatial grid clusters, eliminate discrete noise points, and perform spatial range vectorization encapsulation on the grid clusters that meet the verification to generate a spatial region candidate set.

4. The method according to claim 1, characterized in that, S3 includes: S31: Perform vertical integration and smoothing on the multi-level wind field data in the spatiotemporal data cube, calculate the guiding airflow field representing the average delivery path of the catalyst, and generate the driving wind field for particle diffusion simulation. S32: Based on the driving wind field, perform Lagrange particle reverse trajectory simulation processing, release a large number of virtual particles from the work point, and calculate the three-dimensional backward trajectory and spatial residence probability of the virtual particles within a certain period of time based on the wind field data to generate the potential influence range probability field of the catalyst. S33: Perform spatial overlay and overlap analysis on the potential influence range probability field and the spatial region candidate set, calculate the spatial fit between each candidate region and the high-value region of the potential influence range probability field, and determine the candidate region with the highest spatial fit as the optimal target region. The formula for calculating the spatial fit is: in, Candidate region Space adaptability For grid points The probability value of the catalyst at that location. This refers to the high-probability region comprised of all grid points in the probability field whose probability values ​​exceed a preset probability threshold. Spatial weights related to the distance from the grid point to the work point and the terrain uplift effect, Candidate region The total number of grid points contained.

5. The method according to claim 1, characterized in that, S4 includes: S41: After excluding the optimal target area from the candidate set of spatial regions, representative feature extraction processing is performed on the remaining candidate regions. The static feature mean vector and the change sequence of dynamic meteorological features before and after the operation are calculated for each region to generate a standardized feature profile of each candidate region. S42: Perform multi-dimensional similarity calculations on the standardized feature profiles of each candidate region and the corresponding features of the optimal target region. Comprehensively evaluate the degree of matching between the two in terms of climate background, topographic attributes, and the temporal evolution of key meteorological indicators to generate a comprehensive similarity score for each candidate region relative to the target region. The formula for calculating the comprehensive similarity score is as follows: in, The score represents the combined similarity between candidate region i and the optimal target region T. and These are the static background feature vectors of the target region T and the candidate region i, respectively. Let be the variance of the distances between the static feature vectors of all candidate regions. and These are time series of key dynamic meteorological indicators for the target area T and candidate area i within a preset time period before the operation. For all candidate regions and target region T The maximum distance, and Weighting coefficients to adjust the contributions of static and dynamic similarity; S43: Compare, sort, and filter the comprehensive similarity scores of all candidate regions, select the candidate region with the highest score that exceeds the preset similarity threshold, and generate the optimal comparison region.

6. The method according to claim 1, characterized in that, S5 includes: S51: Based on the spatial boundaries of the optimal target area and the optimal comparison area, perform spatial mask extraction processing on the spatiotemporal data cube, and extract the precipitation rate, radar combined reflectivity and other index data of all grid points in the two areas during the baseline period before the operation and the evaluation period after the operation, respectively, to generate the original index spatiotemporal dataset of the optimal target area and the optimal comparison area. S52: Perform time series aggregation on the original index spatiotemporal dataset, calculate the average index values ​​of the optimal target area and the optimal comparison area at each evaluation time, and remove the average state value of each value from the baseline period before the operation to generate the index increment sequence after background subtraction of the optimal target area and the optimal comparison area. S53: Perform time-by-time difference processing on the index increment sequence, calculate the difference between the index increment of the optimal target area and the index increment of the optimal comparison area at the same time, and generate a time series of catalytic effect difference that characterizes the net effect of catalytic operation.

7. The method according to any one of claims 1-6, characterized in that, S6 includes: S61: Perform time-weighted integral processing based on physical processes on the time series of the catalytic effect difference, assign different contribution weights to the values ​​at different times in the series according to the typical time period characteristics of the occurrence and development of catalytic effect, and accumulate the weighted values ​​to generate the integrated effect value. S62: Perform statistical comparison and analysis on the integrated effect value and the preset significance threshold, determine whether the integrated effect value exceeds the critical level defined by the preset significance threshold, and generate a qualitative judgment result on whether the work effect is significant. S63: Based on the qualitative judgment results, perform conclusion organization and output processing, integrate the qualitative judgment results and the integrated effect value into a structured evaluation statement, and generate an operation effect evaluation conclusion. The operation effect evaluation conclusion is used to indicate whether the artificial rain enhancement operation produces a statistically significant catalytic effect.

8. A system for evaluating the effectiveness of artificial rain enhancement operations based on the index threshold method, characterized in that, The system includes: The spatiotemporal data standardization and fusion module is used to standardize and fuse raw data containing geographical background of the work area, real-time meteorological observations and numerical forecasts, and to align the raw data with spatiotemporal references and unify the format to generate a globally standardized spatiotemporal data cube. The multidimensional feature region candidate generation module is used to extract multidimensional features from the spatiotemporal data cube, automatically classify the grid points based on the climate background, terrain and real-time meteorological features, and generate multiple spatial region candidate sets with similar internal features that can be used as potential analysis units. The optimal target area screening module is used to simulate the transport trajectory of the catalyst in the target cloud system based on the wind field data in the spatiotemporal data cube and the Lagrange particle diffusion model, and to select the area with the highest overlap with the simulated diffusion path from the candidate set of spatial regions to generate the optimal target area. The optimal comparison area filtering module is used to calculate and filter out another area with the highest comprehensive similarity to the optimal target area in terms of climate background and real-time meteorological characteristics from the candidate set of spatial areas, and generate the optimal comparison area. The catalytic effect difference sequence generation module is used to extract precipitation and cloud microphysical index sequences before and after the operation from the spatiotemporal data cube based on the optimal target area and the optimal comparison area, calculate the index time series of the optimal target area and the optimal comparison area respectively and perform background value subtraction to generate a catalytic effect difference time series characterizing the net effect of the catalytic operation. The task effectiveness evaluation module is used to perform integrated calculations on the time series of the catalytic effect difference, obtain the integrated effect value through time-weighted integration, and compare the integrated effect value with the significance threshold obtained based on historical data statistics to generate a task effectiveness evaluation conclusion.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images