Acoustic rainfall enhancement effect evaluation method and system for reservoir storage
By combining topographic elevation data, historical rainfall timing data and satellite cloud map, the similarity between the target area and the candidate area is calculated, the comparison area and time period are determined, and the reservoir accumulation effect of sound wave rain increase operation is evaluated, which solves the accuracy of evaluation in the existing technology The problem of insufficient results has achieved a more accurate assessment of the reservoir accumulation effect.
Patent Information
- Application Number
- CN202411282469.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-13
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-13
AI Technical Summary
The prior art lacks effective methods to evaluate the effectiveness of acoustic rain-enhancing operations aimed at reservoir storage, especially when considering terrain and cloud states, the accuracy of existing methods is insufficient.
By obtaining the topographic elevation data, historical rainfall timing data and satellite cloud map of the target area and candidate area, calculate the similarity between the target area and the candidate area, determine the candidate area that is most similar to the target area as the comparison area, and select the most through the rainfall timing data Similar periods are used to evaluate the reservoir accumulation effect of sonic rain-enhancing operations.
By comprehensively considering the terrain, cloud state and rainfall timing data, this method improves the accuracy of evaluating the effect of sound wave rain increase operation, reduces the influence of human factors, and can more accurately reflect the impact of sound wave rain increase on reservoir storage.
Smart Images

Figure CN118780650B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method and system for evaluating the effect of an acoustic wave rainfall enhancement operation for increasing water storage in a reservoir, and belongs to the technical field of rainfall enhancement effect evaluation. Background Art
[0002] Artificial rainfall technology refers to artificially promoting the rapid condensation or collision of cloud droplets into raindrops, forming rainfall or increasing rainfall. Artificial rainfall technology has the functions of agricultural drought resistance, reservoir storage, air purification, etc. The evaluation of artificial rainfall effect plays an important role in the improvement of specific business and artificial rainfall methods and technologies.
[0003] Acoustic precipitation enhancement uses the energy of sound waves to create tiny vibrations or disturbances in the clouds. These disturbances can increase the frequency and efficiency of collisions between water droplets or ice crystals in the clouds, thereby accelerating the growth of water droplets. When the water droplets grow large enough and become heavy enough, they will fall from the clouds as precipitation.
[0004] The existing artificial rainfall enhancement effect evaluation methods include statistical tests, physical tests and other means, but there is a lack of methods for evaluating the effectiveness of acoustic rainfall enhancement operations aimed at increasing reservoir storage.
[0005] The paper "Evaluation of the Effect of Acoustic Precipitation Enhancement in the Yellow River Source Area Based on Rainfall Temporal Structure Analysis" (Journal of Basic Science and Engineering, 2020, 28(03): 691-702) proposed a method for structuring rainfall data and temporal structure analysis methods such as two-stage decomposition, four-stage decomposition and minimum temporal resolution decomposition of rainfall time series. This method is aimed at evaluating the effect of acoustic precipitation enhancement in agricultural drought resistance. For acoustic precipitation enhancement for the purpose of increasing reservoir storage, the evaluation results are not accurate. Summary of the invention
[0006] In order to overcome the above problems, the present invention provides a method and system for evaluating the effect of acoustic rainfall enhancement operations for reservoir storage.
[0007] The technical solution of the present invention is as follows:
[0008] First aspect
[0009] A method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage, comprising the following steps:
[0010] Delineating a target area for acoustic rainfall enhancement according to the content of the acoustic rainfall enhancement operation to be evaluated, wherein the target area is a rectangle, and delineating a number of candidate areas with the same shape and area outside the target area;
[0011] Obtain terrain elevation data of target area and candidate area;
[0012] Obtain historical rainfall time series data collected by several rain gauges in the target area and candidate areas;
[0013] Obtain satellite cloud images of the target area and candidate area before and after the acoustic rain enhancement operation;
[0014] Calculate the similarity between the target area and each candidate area according to the historical rainfall time series data, the terrain elevation data and the satellite cloud image, and use the candidate area with the highest similarity to the target area as the comparison area;
[0015] Obtain historical rainfall time series data for the target area;
[0016] Calculate the similarity between the rainfall time series data of the comparison area and the historical rainfall time series data of the target area before and after the acoustic rainfall enhancement operation, and take the period corresponding to the historical rainfall time series data with the highest similarity as the comparison period;
[0017] The reservoir storage increase effect of this acoustic rainfall enhancement operation is evaluated based on the reservoir storage increase of the acoustic rainfall enhancement operation to be evaluated and the reservoir storage increase of the comparison period.
[0018] Furthermore, the historical rainfall time series data collected by several rain gauges in the target area and candidate area are obtained, specifically:
[0019] Get the target area n Historical rainfall time series data collected by rain gauge stations :
[0020] ;
[0021] in, Target area n The rainfall station M The amount of rainfall collected during each period;
[0022] Get the l The candidate district n Historical rainfall time series data collected by rain gauge stations :
[0023] ;
[0024] in, For the l The candidate district n The rainfall station M The amount of rainfall collected during each period;
[0025] The historical rainfall time series data and the historical rainfall time series data collected during the same rainfall;
[0026] The historical rainfall time series data and historical rainfall time series data As a set of historical rainfall time series data;
[0027] Repeat the above steps to obtain several sets of historical rainfall time series data.
[0028] Furthermore, the historical rainfall time series data Obtained by:
[0029] According to historical rainfall time series data Get the historical rainfall time series data of the target area :
[0030] ;
[0031] in, The target area is i The amount of rainfall in a period;
[0032] ;
[0033] in, Target area N The rainfall station i The rainfall collected during the period, N is the number of rainfall stations in the target area;
[0034] Get the l The candidate district n Historical rainfall time series data collected by rain gauge stations :
[0035] ;
[0036] in, For the l The candidate district n The rainfall station The rainfall collected during the period, Greater than M ;
[0037] According to historical rainfall time series data Get the first l Historical rainfall time series data for candidate areas :
[0038] ;
[0039] in, For the l The candidate district is i The amount of rainfall in a period;
[0040] ;
[0041] in, is the number of rainfall stations in the candidate area;
[0042] From the historical rainfall time series data Extract some historical rainfall time series data:
[0043] ;
[0044] in, The historical rainfall time series data The first j Historical rainfall time series data, For the l The candidate district is j The rainfall in a period of time, ;
[0045] According to the historical rainfall time series data Calculate total rainfall in the target area;
[0046] According to the historical rainfall time series data Calculate the l Total rainfall in each candidate area;
[0047] Eliminate historical rainfall time series data whose total rainfall does not meet the preset conditions ;
[0048] Calculate the historical rainfall time series data Compared with historical rainfall time series data The similarity will be with the historical rainfall time series data The rainfall time series data with the highest similarity The corresponding time period is l Collect historical rainfall time series data for each candidate area period of time.
[0049] Further, according to the historical rainfall time series data, the terrain elevation data and the satellite cloud image, the similarity between the target area and each candidate area is calculated, and the candidate area with the highest similarity to the target area is used as the comparison area, including:
[0050] The target area and the candidate area are divided into a plurality of non-overlapping rectangular areas, and the target area and the candidate area are divided in the same way, and each of the rectangular areas contains a rainfall station;
[0051] Converting a set of historical rainfall time series data of the rectangular area into image A through the Gram angle field;
[0052] Acquire historical time-series rainfall images of the target area and each candidate area, wherein the historical time-series rainfall images include a plurality of rectangular areas, and each rectangular area is stretched and filled by the corresponding image A;
[0053] Calculate the similarity between the historical time series rainfall image of the target area and the historical time series rainfall images of each candidate area to obtain a first similarity;
[0054] Repeat the above steps to calculate the first similarity of each group of historical rainfall time series data between the target area and each candidate area;
[0055] A second similarity between the target area and each candidate area is obtained, where the second similarity is obtained by taking a median value of each of the first similarities.
[0056] Further, according to the rainfall time series data, the terrain elevation data and the satellite cloud image, the similarity between the target area and each candidate area is calculated, and the candidate area with the highest similarity to the target area is used as the comparison area, which also includes:
[0057] Converting the terrain elevation data of the target area and the candidate area into a grayscale image B;
[0058] Converting the satellite cloud images of the target area and the candidate area into grayscale images C;
[0059] Superimposing the image B and the image C to obtain image D;
[0060] Extracting features related to rainfall in the image D by a feature extraction module to obtain a feature map A;
[0061] Calculate the similarity between the target area feature map A and each candidate area feature map A to obtain a third similarity;
[0062] The comprehensive similarity between the target area and each candidate area is calculated, where the comprehensive similarity is obtained by weighting the second similarity and the third similarity.
[0063] Furthermore, the feature extraction module includes a convolution layer, a pooling layer, a Dense Block layer, a global average pooling layer and an output layer connected in sequence;
[0064] The convolutional layer contains a 7×7 convolution kernel with a stride of 2;
[0065] The pooling window of the pooling layer is 3×3 and the step size is 2;
[0066] The Dense Block layer includes 4 sequentially connected convolution blocks, each convolution block includes 4 convolution layers, each convolution layer is connected to a batch normalization layer and an activation function, and the convolution blocks are connected through a Transition Block layer;
[0067] The Transition Block layer includes a convolution layer with a convolution kernel of 1×1 and an average pooling layer with a pooling window of 2×2.
[0068] Furthermore, the similarity between the feature map A of the target area and each candidate area is calculated to obtain a third similarity, which is specifically:
[0069] The third similarity between the target area and the feature map A of each candidate area is calculated by the normalized cross-correlation matching algorithm.
[0070] Furthermore, it also includes the evaluation of the effect of acoustic rainfall enhancement, specifically:
[0071] ;
[0072] in, is the rainfall in the target area on the day of the acoustic rainfall enhancement operation, The rainfall in the comparison area on the day of the acoustic rainfall enhancement operation. For the acoustic rain enhancement operation i Daily rainfall in the comparison area, For the acoustic rainfall enhancement operation i Daily rainfall in target area.
[0073] Second aspect
[0074] A system for evaluating the effect of acoustic wave rainfall enhancement operations for increasing water storage in a reservoir comprises a storage unit and a data processing unit. The data processing unit executes the method for evaluating the effect of acoustic wave rainfall enhancement operations for increasing water storage in a reservoir as described in the first aspect to evaluate the effect of the acoustic wave rainfall enhancement operations.
[0075] The third aspect
[0076] A storage medium, which, when the computer program stored in the storage medium is executed, implements the method for evaluating the effect of acoustic rainfall enhancement operations for reservoir storage as described in the first aspect.
[0077] The present invention has the following beneficial effects:
[0078] The state, position, movement direction of clouds and ground topography have a great influence on rainfall, and the combination of clouds and terrain also has an impact on rainfall. The existing technology usually only considers the rainfall of artificial rainfall, but ignores the time series data of rainfall. The time series data of rainfall can reflect the state of clouds. According to the time series data of rainfall, the comparison area can be better confirmed. This method integrates terrain elevation data and satellite cloud images, extracts features related to clouds, terrain and rainfall through the feature extraction module, and calculates the similarity between the target area and the candidate area through the feature. The similarity between the target area and the candidate area is calculated by the time series data of rainfall, and finally the comparison area is determined by combining the two. Compared with the existing technology, more factors are considered.
[0079] After determining the rainfall period of the comparison area, the prior art usually artificially limits the rainfall period of the candidate area to perform similarity calculation. Artificially limiting the candidate area causes the similarity calculation result to be greatly affected by subjective factors and is inaccurate. This method compares the rainfall time series data with the rainfall time series data of several candidate time periods, and selects the rainfall period with the highest similarity as the candidate area. This effectively improves the accuracy of the candidate area rainfall period selection.
[0080] The existing technology usually evaluates the effect of rainfall enhancement. The purpose of this method is to evaluate the water storage effect after acoustic rainfall enhancement. Therefore, only by comparing the rainfall in the target area and the comparison area before and after the acoustic rainfall enhancement to evaluate the operation effect, it cannot accurately reflect the impact of the reservoir's rainfall enhancement operation on the reservoir's storage. This method compares the rainfall conditions in the comparison area before and after the acoustic rainfall enhancement operation to be evaluated with the historical rainfall time series data of the target area, and selects the time period of the close historical rainfall time series data as the comparison period. Therefore, the rainfall conditions in the target area during the comparison period are similar to the rainfall conditions in the target area before and after the acoustic rainfall enhancement operation to be evaluated. At this time, the evaluation of the reservoir's storage capacity is most accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0081] Figure 1 The figure is a flow chart of the evaluation method of the present invention.
[0082] Figure 2 These are the target area and candidate area in the embodiment of the present invention.
[0083] Figure 3 It is a schematic diagram of the rectangular area division of the target area and the candidate area according to an embodiment of the present invention.
[0084] Figure 4 Schematic diagram of the feature extraction module structure of an embodiment of the present invention. DETAILED DESCRIPTION
[0085] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0086] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings of the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0087] Unless otherwise defined, the technical terms or scientific terms used in the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0088] The existing technology selects the comparison area through historical rainfall data, which does not take into account the impact of terrain on rainfall. In actual operation, the terrain has a great impact on rainfall, especially in mountainous areas. The existing technology only evaluates the rainfall increase of acoustic rain enhancement, which has a good effect in the fields of agricultural drought resistance, but for acoustic rain enhancement operations aimed at increasing reservoir storage, the rainfall increase can only be used as an indirect indicator, and the evaluation result is not accurate.
[0089] The present invention overcomes the above problems by the following method.
[0090] First aspect
[0091] refer to Figure 1 and 2 A method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage includes the following steps:
[0092] Delineating a target area for acoustic rainfall enhancement according to the content of the acoustic rainfall enhancement operation to be evaluated, wherein the target area is a rectangle, and delineating a number of candidate areas with the same shape and area outside the target area;
[0093] Obtain terrain elevation data of target area and candidate area;
[0094] Obtain historical rainfall time series data collected by several rain gauges in the target area and candidate areas;
[0095] Obtain satellite cloud images of the target area and candidate area before and after the acoustic rain enhancement operation;
[0096] Calculate the similarity between the target area and each candidate area according to the historical rainfall time series data, the terrain elevation data and the satellite cloud image, and use the candidate area with the highest similarity to the target area as the comparison area;
[0097] Obtain historical rainfall time series data for the target area;
[0098] Calculate the similarity between the rainfall time series data of the comparison area and the historical rainfall time series data of the target area before and after the acoustic rainfall enhancement operation, and take the period corresponding to the historical rainfall time series data with the highest similarity as the comparison period;
[0099] The reservoir storage increase effect of this acoustic rainfall enhancement operation is evaluated based on the reservoir storage increase of the acoustic rainfall enhancement operation to be evaluated and the reservoir storage increase of the comparison period.
[0100] Figure 2 The black framed area is the target area, and the white framed area is the candidate area.
[0101] This method calculates the similarity between the target area and each candidate based on historical rainfall time series data, terrain elevation data and satellite cloud images, and uses the candidate area with the highest similarity to the target area as the comparison area.
[0102] The similarity between the rainfall time series data of the comparison area and the historical rainfall time series data of the target area is then calculated to determine which period of rainfall in the target area's history is most similar to the current rainfall in the comparison area.
[0103] At this time, the rainfall conditions in the target area during the comparison period should be equivalent to those in the target area without acoustic rainfall enhancement.
[0104] Therefore, directly comparing the storage effect of the comparison period with the storage effect of this acoustic rainfall enhancement operation can most accurately evaluate the effect of acoustic rainfall enhancement.
[0105] In one embodiment of the present invention, historical rainfall time series data collected by several rain gauges in the target area and the candidate area are obtained, specifically:
[0106] Get the target area n Historical rainfall time series data collected by rain gauge stations :
[0107] ;
[0108] in, Target area n The rainfall station M The amount of rainfall collected during each period;
[0109] Get the l The candidate district n Historical rainfall time series data collected by rain gauge stations :
[0110] ;
[0111] in, For the l The candidate district nThe rainfall station M The amount of rainfall collected during each period;
[0112] The historical rainfall time series data and the historical rainfall time series data collected during the same rainfall;
[0113] The historical rainfall time series data and historical rainfall time series data As a set of historical rainfall time series data;
[0114] Repeat the above steps to obtain several sets of historical rainfall time series data.
[0115] When selecting the comparison area, it is necessary to calculate the similarity between the historical rainfall data of each candidate area and the target area, and select the candidate area with the highest similarity as the comparison area.
[0116] The present invention uses rainfall time series data, so it is necessary to select rainfall time series data with the same length. A set of rainfall time series data is collected from the same rainfall, and includes rainfall time series data of each candidate area in the target area in the rainfall.
[0117] When comparing the candidate area with the target area, it is necessary to confirm the comparison period. Since the rainfall time in different areas is not exactly the same, it is necessary to select the rainfall period separately. The existing technology usually selects the rainfall period manually, and the result is too much affected by human subjective factors. Since this method uses the time series data of rainfall to find the comparison area, the selection of the rainfall period is more important.
[0118] In one embodiment of the present invention, the historical rainfall time series data Obtained by:
[0119] According to historical rainfall time series data Get the historical rainfall time series data of the target area :
[0120] ;
[0121] in, The target area is i The amount of rainfall in a period;
[0122] ;
[0123] in, Target area N The rainfall station i The rainfall collected during the period, N is the number of rainfall stations in the target area, that is, iThe rainfall in a period is determined by the average of all rainfall stations in the target area during that period;
[0124] Get the l The candidate district n Historical rainfall time series data collected by rain gauge stations :
[0125] ;
[0126] in, For the l The candidate district n The rainfall station The rainfall collected during the period, Greater than M ;
[0127] According to historical rainfall time series data Get the first l Historical rainfall time series data for candidate areas :
[0128] ;
[0129] in, For the l The candidate district is i The amount of rainfall in a period;
[0130] ;
[0131] in, is the number of rainfall stations in the candidate area, that is, l The candidate district is i The rainfall in a period is given by l The mean value of all rainfall stations in the candidate area is determined;
[0132] From the historical rainfall time series data Extract some historical rainfall time series data:
[0133] ;
[0134] in, The historical rainfall time series data The first j Historical rainfall time series data, For the l The candidate district is j The rainfall in a period of time, ;
[0135] According to the historical rainfall time series data Calculate total rainfall in the target area;
[0136] According to the historical rainfall time series data Calculate the l Total rainfall in each candidate area;
[0137] Eliminate historical rainfall time series data whose total rainfall does not meet the preset conditions ;
[0138] Calculate the historical rainfall time series data Compared with historical rainfall time series data The similarity will be with the historical rainfall time series data The rainfall time series data with the highest similarity The corresponding time period is l Collect historical rainfall time series data for each candidate area period of time.
[0139] The method first obtains a candidate area rainfall time series data that is longer than the target area rainfall time series data. The candidate area rainfall time series data is selected from a rain that is the same as the target area rainfall time series data, and can usually be selected from 5 hours before the start time of the target area rainfall time series data to 5 hours after the end time.
[0140] From this longer candidate area rainfall time series data, a number of data with the same length as the target area rainfall time series data are selected, and the similarity between the target area rainfall time series data and the rainfall time series data of each selected area is calculated, so as to obtain the most similar rainfall period in the candidate area of the target area in a rain, thereby improving the accuracy of rainfall period selection.
[0141] refer to Figure 3 In one embodiment of the present invention, the similarity between the target area and each candidate area is calculated based on the historical rainfall time series data, the terrain elevation data and the satellite cloud image, and the candidate area with the highest similarity to the target area is used as the comparison area, including:
[0142] The target area and the candidate area are divided into a plurality of non-overlapping rectangular areas, and the target area and the candidate area are divided in the same way, and each of the rectangular areas contains a rainfall station;
[0143] Converting a set of historical rainfall time series data of the rectangular area into image A through the Gram angle field;
[0144] Acquire historical time-series rainfall images of the target area and each candidate area, wherein the historical time-series rainfall images include a plurality of rectangular areas, and each rectangular area is stretched and filled by the corresponding image A;
[0145] Calculate the similarity between the historical time series rainfall image of the target area and the historical time series rainfall images of each candidate area to obtain a first similarity;
[0146] Repeat the above steps to calculate the first similarity of each group of historical rainfall time series data between the target area and each candidate area;
[0147] A second similarity between the target area and each candidate area is obtained, where the second similarity is obtained by taking a median value of each of the first similarities.
[0148] The Gram angle field can convert time series into images, fill the time images collected by each rain gauge station in the corresponding rectangular area, and obtain images related to the candidate area and rainfall time series data.
[0149] Since the rectangular area division method of the target area and the candidate area is the same, the rainfall time series data of each rectangular area can be taken into account when performing image similarity matching. Figure 3 This is a schematic diagram of rectangular area division according to an embodiment of the present invention, where each rectangular area includes a rainfall station.
[0150] Meteorological data and topographic data are both very important for rainfall, and the combination of the two has an impact on rainfall. Therefore, when considering the situation of this acoustic rainfall enhancement operation, it is necessary to consider both of them comprehensively.
[0151] In one embodiment of the present invention, the similarity between the target area and each candidate is calculated based on the rainfall time series data, the terrain elevation data and the satellite cloud image, and the candidate area with the highest similarity to the target area is used as the comparison area, which also includes:
[0152] Converting the terrain elevation data of the target area and the candidate area into a grayscale image B;
[0153] Converting the satellite cloud images of the target area and the candidate area into grayscale images C;
[0154] Superimposing the image B and the image C to obtain image D;
[0155] Extracting features related to rainfall in the image D by a feature extraction module to obtain a feature map A;
[0156] Calculate the similarity between the target area feature map A and each candidate area feature map A to obtain a third similarity;
[0157] The comprehensive similarity between the target area and each candidate area is calculated, where the comprehensive similarity is obtained by weighting the second similarity and the third similarity.
[0158] In a specific embodiment, the image B and the image C are superimposed by making the image B and the image C 50% transparent respectively and then superimposing them.
[0159] DenseNet is a convolutional neural network architecture that has good effects in image feature extraction.
[0160] refer to Figure 4 In one embodiment of the present invention, the feature extraction module includes a convolution layer, a pooling layer, a Dense Block layer, a global average pooling layer and an output layer connected in sequence;
[0161] The convolutional layer contains a 7×7 convolution kernel with a stride of 2;
[0162] The pooling window of the pooling layer is 3×3 and the step size is 2;
[0163] The Dense Block layer includes 4 sequentially connected convolution blocks, each convolution block includes 4 convolution layers, each convolution layer is connected to a batch normalization layer and an activation function, and the convolution blocks are connected through a Transition Block layer;
[0164] The Transition Block layer includes a convolution layer with a convolution kernel of 1×1 and an average pooling layer with a pooling window of 2×2.
[0165] When evaluating the effect of artificial rainfall enhancement operations, the target area is manually delineated. Conventional evaluations can be carried out according to the needs of the target area, mainly considering the difficulty of data collection and processing and the scope of influence of the rainfall enhancement operation. In the artificial sound wave rainfall enhancement operation used for reservoir storage, the division of the target area is more complicated, because the reservoir itself is large in area and irregular in shape, and the storage of some reservoirs does not come directly from rainfall in the reservoir surface area, but from the influx of water. Therefore, when delineating the target area, it is usually necessary to delineate a larger range.
[0166] Since the resolution of terrain elevation data and satellite cloud images is high and the demarcation range is too large, it will lead to difficulties in data processing. Therefore, the data needs to be downsampled to reduce the resolution.
[0167] In addition to the downsampling process, the method further reduces the image size through the convolution layer and the pooling layer.
[0168] The similarity between the target area feature map A and each candidate area feature map A is calculated to obtain a third similarity, and the image similarity can be calculated using corner point features.
[0169] In one embodiment of the present invention, the similarity between the feature map A of the target area and each candidate area is calculated to obtain a third similarity, which is specifically:
[0170] The third similarity between the target area and the feature map A of each candidate area is calculated by the normalized cross-correlation matching algorithm.
[0171] In one embodiment of the present invention, the effect of acoustic rainfall enhancement is also evaluated, specifically:
[0172] ;
[0173] in, is the rainfall in the target area on the day of the acoustic rainfall enhancement operation, The rainfall in the comparison area on the day of the acoustic rainfall enhancement operation. For the acoustic rainfall enhancement operation i Daily rainfall in the comparison area, For the acoustic rainfall enhancement operation i Daily rainfall in target area.
[0174] In one embodiment of the present invention, the similarity between the rainfall time series data of the comparison area before and after the acoustic rainfall enhancement operation and the historical rainfall time series data of the target area is calculated, and the time period corresponding to the historical rainfall time series data with the highest similarity is used as the comparison period, specifically:
[0175] The target area and the comparison area are re-divided into a number of non-overlapping rectangular areas, and the division method of the target area and the candidate area is the same, each of the rectangular areas contains a rain gauge, and the rain gauge is close to the center of the rectangle;
[0176] The rainfall time series data of the comparison area before and after the acoustic rainfall enhancement operation and the historical rainfall time series data of the target area are converted into image E through the Gram Corner Field;
[0177] Acquire each historical time series rainfall image of the target area and the rainfall image of the comparison area before and after the acoustic rain enhancement operation, wherein the rainfall image includes a plurality of rectangular areas, and each of the rectangular areas is stretched and filled by the corresponding image E;
[0178] The similarity between the rainfall images before and after the acoustic rainfall enhancement operation in the comparison area and the historical time series rainfall images in the target area is calculated, and the period corresponding to the historical time series rainfall image with the highest similarity is taken as the comparison period.
[0179] When selecting the comparison area, the target area and the candidate area should be divided into the same rectangular area. Due to the different distribution positions of the rain gauges, the rain gauges contained in each candidate area and the target area should be considered when dividing the rectangular area. The number of divided rectangular areas is small to ensure that there are rain gauges in the rectangular areas of each candidate area and the target area. And at this time, the rain gauge is not necessarily located in the center of the rectangular area. Therefore, when calculating the similarity of the image, it cannot reflect the distribution of rainfall very well.
[0180] In this step, the target area and the comparison area are re-divided into rectangular areas. As many rectangular areas as possible can be divided according to the distribution of rain gauges, and the rain gauges are placed as close to the center of the rectangle as possible.
[0181] This method can better reflect the distribution of rainfall.
[0182] Second aspect
[0183] A system for evaluating the effect of acoustic wave rainfall enhancement operations for increasing water storage in a reservoir comprises a storage unit and a data processing unit. The data processing unit executes the method for evaluating the effect of acoustic wave rainfall enhancement operations for increasing water storage in a reservoir described in the first aspect to evaluate the effect of the acoustic wave rainfall enhancement operations.
[0184] The third aspect
[0185] A storage medium, which, when the computer program stored in the storage medium is executed, implements the method for evaluating the effect of acoustic rainfall enhancement operations for reservoir storage as described in the first aspect.
[0186] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure made using the contents of the present invention's specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage, characterized in that: The following steps are involved: Delineating a target area for acoustic rainfall enhancement according to the content of the acoustic rainfall enhancement operation to be evaluated, wherein the target area is a rectangle, and delineating a number of candidate areas with the same shape and area outside the target area; Obtain terrain elevation data of target area and candidate area; Obtain historical rainfall time series data collected by several rain gauges in the target area and candidate areas; Obtain satellite cloud images of the target area and candidate area before and after the acoustic rainfall enhancement operation; Calculate the similarity between the target area and each candidate area according to the historical rainfall time series data, the terrain elevation data and the satellite cloud image, and use the candidate area with the highest similarity to the target area as the comparison area; Obtain rainfall time series data in the comparison area before and after the acoustic rainfall enhancement operation; Calculate the similarity between the rainfall time series data of the comparison area and the historical rainfall time series data of the target area before and after the acoustic rainfall enhancement operation, and take the period corresponding to the historical rainfall time series data with the highest similarity as the comparison period; Evaluate the reservoir storage effect of the acoustic rainfall enhancement operation based on the reservoir storage increase of the acoustic rainfall enhancement operation to be evaluated and the reservoir storage increase of the comparison period; Obtain the historical rainfall time series data collected by several rain gauges in the target area and candidate area, specifically: Obtain the historical rainfall time series data A collected by the nth rainfall station in the target area n : A n =(a n1 ,a n2 ,…,a nM ); Among them, a nM is the rainfall collected by the nth rainfall station in the target area in the Mth period; Get the historical rainfall time series data B collected by the nth rainfall station in the lth candidate area ln : B ln =(b ln1 ,b ln2 ,…,b lnM ); Among them, b lnM is the rainfall collected by the nth rainfall station in the lth candidate area in the Mth period; The historical rainfall time series data A n and the historical rainfall time series data B ln collected during the same rainfall; The historical rainfall time series data A n and historical rainfall time series data B ln As a set of historical rainfall time series data; Repeat the above steps to obtain several sets of historical rainfall time series data; The historical rainfall time series data B ln Obtained by: According to the historical rainfall time series data A n Get the historical rainfall time series data A0 of the target area: A0=(a 01 ,to 02 ,…,to 0i …,to 0M ): Among them, a 0i is the rainfall in the target area in the i-th period; Among them, a Ni is the rainfall collected by the Nth rainfall station in the target area in the i-th period, and N is the number of rainfall stations in the target area; Get the historical rainfall time series data C collected by the nth rainfall station in the lth candidate area ln : C ln =(c ln1 ,c ln2 ,…,c lnM′ ); Among them, c lnM ′ is the rainfall collected by the nth rainfall station in the lth candidate area in the M′th period, and M′ is greater than M; According to the historical rainfall time series data C ln Get the historical rainfall time series data C of the lth candidate area l0 : C l0 =(c l01 ,c l02 ,…,c l0i ,…,c l0M′ ); Among them, c l0i is the rainfall of the lth candidate area in the i-th period; Where N′ is the number of rainfall stations in the candidate area; From the historical rainfall time series data C l0 Extract some historical rainfall time series data: C l0j =(c l0j ,c l0(j+1) ,…,c l0(j+M-1) ) Among them, C l0j is the historical rainfall time series data C l0 The jth historical rainfall time series data extracted from l0j is the rainfall of the lth candidate area in the jth period, j∈[1,M′-M+1]; Calculate the total rainfall in the target area according to the historical rainfall time series data A0; According to the historical rainfall time series data C l0j Calculate the total rainfall in the lth candidate area; Eliminate the historical rainfall time series data C where the total rainfall does not meet the preset conditions l0j ; Calculate the historical rainfall time series data A0 and the historical rainfall time series data C l0j The rainfall time series data C with the highest similarity to the historical rainfall time series data A0 l0j The corresponding period is taken as the lth candidate area to collect historical rainfall time series data B ln period of time.
2. The method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage according to claim 1 is characterized in that: According to the historical rainfall time series data, the terrain elevation data and the satellite cloud map, the similarity between the target area and each candidate area is calculated, and the candidate area with the highest similarity to the target area is used as the comparison area, including: The target area and the candidate area are divided into a number of non-overlapping rectangular areas, and the target area and the candidate area are divided in the same way, and each of the rectangular areas contains a rainfall station; Converting a set of historical rainfall time series data of the rectangular area into image A through the Gram angle field; Acquire historical time-series rainfall images of the target area and each candidate area, wherein the historical time-series rainfall images include a plurality of rectangular areas, and each rectangular area is stretched and filled by the corresponding image A; Calculate the similarity between the historical time series rainfall image of the target area and the historical time series rainfall images of each candidate area to obtain a first similarity; Repeat the above steps to calculate the first similarity of each group of historical rainfall time series data between the target area and each candidate area; A second similarity between the target area and each candidate area is obtained, where the second similarity is obtained by taking a median value of each of the first similarities.
3. The method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage according to claim 2 is characterized in that: According to the rainfall time series data, the terrain elevation data and the satellite cloud image, the similarity between the target area and each candidate area is calculated, and the candidate area with the highest similarity to the target area is used as the comparison area, further comprising: Converting the terrain elevation data of the target area and the candidate area into a grayscale image B; Converting the satellite cloud images of the target area and the candidate area into grayscale images C; Superimposing the image B and the image C to obtain image D; Extracting features related to rainfall in the image D by a feature extraction module to obtain a feature map A; Calculate the similarity between the target area feature map A and each candidate area feature map A to obtain a third similarity; The comprehensive similarity between the target area and each candidate area is calculated, where the comprehensive similarity is obtained by weighting the second similarity and the third similarity.
4. The method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage according to claim 3 is characterized in that: The feature extraction module includes a convolution layer, a pooling layer, a Dense Block layer, a global average pooling layer and an output layer connected in sequence; The convolutional layer contains a 7×7 convolution kernel with a stride of 2; The pooling window of the pooling layer is 3×3 and the step size is 2; The Dense Block layer includes 4 sequentially connected convolution blocks, each convolution block includes 4 convolution layers, each convolution layer is connected to a batch normalization layer and an activation function, and the convolution blocks are connected through a Transition Block layer; The Transition Block layer includes a convolution layer with a convolution kernel of 1×1 and an average pooling layer with a pooling window of 2×2.
5. The method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage according to claim 3 is characterized in that: The similarity between the target area and the feature map A of each candidate area is calculated to obtain the third similarity, which is specifically: The third similarity between the target area and the feature map A of each candidate area is calculated by the normalized cross-correlation matching algorithm.
6. The method for evaluating the effect of acoustic rainfall enhancement operation for reservoir storage according to claim 1 is characterized in that: It also includes the evaluation of the effect of acoustic rainfall enhancement, specifically: Among them, H t is the rainfall in the target area on the day of acoustic rainfall enhancement operation, H d is the rainfall in the comparison area on the day of acoustic rainfall enhancement operation, Q i is the rainfall in the comparison area on the day before the acoustic rainfall enhancement operation, P i It is the rainfall in the target area on the i-th day before the acoustic rainfall enhancement operation.
7. A system for evaluating the effect of acoustic rainfall enhancement operation for increasing the storage capacity of a reservoir, comprising a storage unit and a data processing unit, characterized in that: The data processing unit executes the method for evaluating the effect of acoustic wave rainfall enhancement operation for reservoir storage as described in any one of claims 1-6 to evaluate the effect of the acoustic wave rainfall enhancement operation.
8. A storage medium, characterized in that: When the computer program stored therein is executed, the method for evaluating the effect of acoustic rainfall enhancement operations for reservoir storage as described in any one of claims 1-6 is implemented.
Citation Information
Patent Citations
Aircraft precipitation enhancement effect evaluation method based on satellite remote sensing data
CN114660678A
