Joint early warning method and device, electronic equipment and storage medium

Abstract

The invention provides a joint early warning method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a plurality of cloud / ground lightning monitoring data collected by at least one lightning locator in a monitoring region, and a plurality of electric field intensity change data collected by at least one atmospheric electric field instrument; after a thunderstorm cloud trajectory communication scheme between thunderstorm cloud areas is obtained according to the cloud / ground lightning monitoring data, a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of a thunderstorm cloud tracking trajectory is constructed; after eliminating a data residual error in each electric field intensity change data, constructing an electric field early warning model, and generating electric field early warning information at least comprising an electric field amplitude value and an electric field difference value according to a preset condition; and performing combined early warning according to the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field early warning information. According to the method, the obtained thunderstorm cloud tracking trajectory, the obtained electric field amplitude early warning result and the obtained electric field difference early warning result are combined, and a more accurate thunder and lightning early warning result can be obtained.

Description

Technical Field

[0001] The present invention relates to an early warning method, in particular to a combined early warning method, device, electronic equipment and storage medium. Background Art

[0002] Lightning can cause casualties, livestock deaths, and house and forest fires. It is also a major source of electromagnetic interference and damage to electronic circuits, buildings, and other exposed man-made structures. However, the charge distribution in the atmosphere is extremely complex and dynamically changing. The generation of lightning depends on the uneven distribution of charge within the cloud or between the cloud and the ground. The formation and evolution of this unevenness is difficult to accurately predict.

[0003] Since thunderstorms have a short life cycle and a small range, the operational numerical forecast model has a small temporal and spatial scalability, low horizontal and vertical resolution, and there will be temporal and spatial deviations. Since the occurrence and development of severe convective weather are sudden and localized, the prediction and warning of lightning within a few hours rely more on real-time monitoring methods. The atmospheric electric field meter is a non-directional detection device that cannot detect the specific location of the thunderstorm and is very sensitive to changes in charge caused by environmental factors. Similarly, the lightning locator data records the lightning events that have already occurred, which is less timely for lightning warning. Therefore, it is considered to conduct a joint warning in the thunderstorm warning area based on the temporal and spatial relationship between the electric field and lightning location data.

[0004] At present, the use of data from multiple devices for thunderstorm warnings, or the combination of real-time observation data with numerical forecasts, often adopts direct fusion of data in space, without effectively processing the data according to the detection characteristics of the equipment to make up for the shortcomings of the warning capabilities. Direct fusion of data is heavily dependent on the data of the equipment with the best observation capabilities and accuracy, while the data of other equipment has extremely limited improvement on the warning results. This cannot effectively solve the time limitations of lightning locators and the space limitations of atmospheric electric field instruments.

[0005] Therefore, how to solve the time limitations of lightning locators and the space limitations of atmospheric electric field meters, and obtain more accurate lightning warning results by using multiple cloud / ground flash monitoring data monitored by lightning locators and multiple electric field strength change data monitored by atmospheric electric field meters, is a technical problem that needs to be solved urgently. Summary of the invention

[0006] In response to the shortcomings existing in the above-mentioned problems, the present invention provides a joint warning method, device, electronic device and storage medium which can conduct joint warning based on multiple cloud / ground flash monitoring data monitored by a lightning locator and multiple electric field strength change data monitored by an atmospheric electric field meter to obtain more prepared lightning warning results.

[0007] To achieve the above object, in a first aspect, the present invention provides a joint early warning method, comprising the following steps:

[0008] According to a preset time threshold, a plurality of cloud / ground lightning monitoring data collected by at least one lightning locator in the monitoring area and a plurality of electric field intensity change data collected by at least one atmospheric electric field meter are obtained;

[0009] After obtaining the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on the cloud / ground lightning monitoring data, a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory is constructed;

[0010] After eliminating the data residuals in each electric field intensity change data, an electric field early warning model is constructed, and electric field early warning information including at least an electric field amplitude and an electric field differential value is generated according to preset conditions;

[0011] A joint warning is issued based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information.

[0012] In one of the embodiments, in the scheme for obtaining thunderstorm cloud trajectory connectivity between thunderstorm cloud areas according to each cloud / ground lightning monitoring data, it includes: obtaining a thunderstorm cloud range map according to thunderstorm coordinate points in each cloud / ground lightning positioning monitoring data, including:

[0013] Obtain the distance value between two adjacent thunderstorm coordinate points in each cloud / ground lightning location monitoring data;

[0014] Perform curve fitting on multiple distance values ​​to obtain a fitting curve;

[0015] In the fitting curve, take the intervals of equal length forward and backward from each point, and calculate the variance, and take the distribution variance to obtain the inflection point of the distance curve;

[0016] The input parameters of the clustering model are obtained by calculating the inflection point of the distance curve;

[0017] Density clustering is performed according to the input parameters of the clustering model to obtain the thunderstorm cloud structure corresponding to the cloud / ground lightning monitoring data;

[0018] The thunderstorm cloud structure is divided using the confidence ellipse or Quickhull algorithm to obtain the thunderstorm cloud range map.

[0019] In one of the embodiments, the scheme for obtaining thunderstorm cloud track connectivity between thunderstorm cloud areas according to each cloud / ground lightning monitoring data also includes:

[0020] The thunderstorm occurrence time sequence state is progressively classified according to the collection time sequence of each cloud / ground lightning monitoring data;

[0021] Calculate the fitness of any two thunderstorm cloud range maps at adjacent times;

[0022] A fitness matrix is ​​constructed based on the multiple fitness measures obtained;

[0023] The Viterbi dynamic programming algorithm or the Hungarian matching algorithm is used to obtain the thunderstorm cloud track connectivity scheme between thunderstorm cloud areas.

[0024] In one of the embodiments, the step of generating electric field warning information at least including an electric field amplitude and an electric field differential value according to a preset condition includes:

[0025] The electric field amplitude is obtained according to the electric field strength change data, and when the electric field amplitude meets the preset conditions, the electric field amplitude warning information is generated;

[0026] An electric field differential value is obtained according to the electric field amplitude, and when the electric field differential value meets a preset condition, an electric field differential value warning information is generated.

[0027] In one embodiment, when the electric field amplitude meets the preset condition, it includes:

[0028] At least comparing the plurality of electric field amplitudes within a preset time threshold with preset amplitude thresholds respectively, and generating electric field amplitude warning information when the preset conditions are met;

[0029] In the case where the electric field difference value meets a preset condition, it includes:

[0030] At least comparing the plurality of electric field differential values ​​within a preset time threshold with preset differential value thresholds, and generating electric field amplitude warning information when a preset condition is met;

[0031] The plurality of prior electric field difference values ​​correspond to the plurality of prior electric field amplitudes.

[0032] In one embodiment, the joint warning based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information also includes:

[0033] After obtaining prior values ​​of multiple historical data in the monitoring area according to the verification indicators, a joint early warning general model is constructed;

[0034] The cloud / ground lightning monitoring data within the preset time threshold obtains the current verification value according to the verification index;

[0035] The current verification value is compared with the prior value and the adjacent value of the selected prior value respectively. If the current verification value is greater than the prior value and the adjacent value of the prior value, the current verification value is used as the prior value to iteratively update the joint warning general model.

[0036] In one of the embodiments, the verification index at least includes verifying the hit rate, false alarm rate and false alarm rate of the historical data and the cloud / ground lightning monitoring data;

[0037] The current verification value, the priori value and the neighboring values ​​of the priori value at least include a hit rate value, an error rate value and a false alarm rate value.

[0038] In one embodiment, it also includes:

[0039] Displays the geographical information of the monitoring area, thunderstorm cloud tracking trajectory, electric field amplitude and electric field differential value.

[0040] In a second aspect, the present invention provides a device for implementing the above-mentioned joint early warning method, comprising:

[0041] Lightning locator, which collects and sends multiple cloud / ground lightning monitoring data in the monitoring area according to the preset time threshold;

[0042] The atmospheric electric field instrument collects and sends multiple electric field intensity change data within the monitoring area according to the preset time threshold;

[0043] The thunderstorm cloud motion trajectory extrapolation module is used to obtain the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on the cloud / ground lightning monitoring data, and then construct a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory;

[0044] The electric field warning module is used to eliminate the data residuals in the electric field strength change data, build an electric field warning model, and generate electric field warning information including at least the electric field amplitude and the electric field differential value according to preset conditions;

[0045] The joint warning module is used to provide joint warning based on the dynamic prediction information of thunderstorm cloud tracking trajectory and electric field warning information.

[0046] In one of the embodiments, it also includes: a display module for displaying the geographical information of the monitoring area, the thunderstorm cloud tracking trajectory, the electric field amplitude and the electric field differential value.

[0047] In a third aspect, the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned joint early warning method.

[0048] In a fourth aspect, the present invention provides a storage medium storing a computer program that can be loaded by a processor and execute the above-mentioned joint early warning method.

[0049] Compared with the prior art, the present invention has one of the following advantages:

[0050] The present invention combines the extrapolated results of the thunderstorm trajectory in the next two hours based on the lightning locator data with the electric field amplitude and differential warning results based on the atmospheric electric field instrument data to obtain a more accurate lightning warning result. Different from the previous method of simply using the fusion of data from different lightning detection equipment, the method adopted by the present invention will give full play to the characteristics of different types of data, and solve the warning time limitation of the lightning locator and the warning space limitation of the atmospheric electric field instrument by combining the lightning locator data extrapolation and the electric field instrument network warning;

[0051] New algorithms are used in thunderstorm identification, trajectory tracking and extrapolation using lightning locator data. The distance inflection point calculation method is used in clustering model parameter selection. Trajectory tracking uses a combination of progressive classification, Viterbi dynamic programming algorithm or Hungarian matching algorithm. Trajectory extrapolation uses the decay trend exponential smoothing method that can gradually correct the results.

[0052] In the electric field warning using the data of atmospheric electric field instrument, the impact of the installation environment and the randomness of thunderstorm cloud shape and charge distribution on the electric field warning is effectively solved by integrating the empirical mode decomposition algorithm and increasing the consistency of the warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flowchart of the method part of the present invention;

[0054] Figure 2 A flowchart of a first embodiment of the method part of the present invention;

[0055] Figure 3 It is a structural block diagram of the device part in the present invention;

[0056] Figure 4 It is a structural block diagram of the first embodiment of the device part of the present invention;

[0057] Figure 5 It is a structural schematic diagram of the electronic device in the present invention;

[0058] Figure 6 It is a warning result diagram obtained during the implementation of the joint warning method or the joint warning device in the present invention.

[0059] The main reference numerals are as follows:

[0060] 1-Lightning locator; 2-Atmospheric electric field meter; 3-Collection and distribution module; 4-Thunderstorm cloud motion trajectory extrapolation module; 41-Thunderstorm cloud range map acquisition module; 42-Thunderstorm cloud trajectory connectivity scheme acquisition module; 43-Thunderstorm cloud motion trajectory extrapolation model construction module; 5-Electric field warning module; 51-Data residual elimination module; 52-Electric field amplitude warning information generation module; 53-Electric field differential warning information generation module; 6-Joint warning module; 7-Display module; 8-CPU; 9-ROM; 10-RAM; 11-I / O interface; 12-Input part; 13-Output part; 14-Storage part; 15-Communication part; 16-Drive; 17-Removable media. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] In the description of the present invention, it is necessary to understand that the terms "upper", "lower", "top surface", "bottom surface", "inside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centrally arranged component at the same time. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0063] like Figure 1 As shown, the present invention provides a joint early warning method, comprising the following steps:

[0064] S1. According to a preset time threshold, obtain a plurality of cloud / ground lightning monitoring data collected by at least one lightning locator in the monitoring area, and a plurality of electric field intensity change data collected by at least one atmospheric electric field meter;

[0065] S2. After obtaining the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on each cloud / ground lightning monitoring data, a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory is constructed;

[0066] S3, after eliminating the data residuals in each electric field intensity change data, construct an electric field warning model, and generate electric field warning information including at least an electric field amplitude and an electric field differential value according to preset conditions;

[0067] S4. Conduct joint warning based on the dynamic prediction information of thunderstorm cloud tracking trajectory and electric field warning information;

[0068] S5. Display the geographical information of the monitoring area, thunderstorm cloud tracking trajectory, electric field amplitude and electric field differential value.

[0069] The present invention combines the extrapolated results of thunderstorm trajectories in the next two hours based on lightning locator data with the electric field amplitude and differential warning results based on atmospheric electric field meter data to obtain more accurate lightning warning results. Different from the previous method of simply integrating data from different lightning detection equipment, the method adopted by the present invention will give full play to the characteristics of different types of data, and solve the warning time limitation of lightning locators and the warning space limitation of atmospheric electric field meters through the combination of lightning locator data extrapolation and electric field meter networking warning.

[0070] Embodiment 1

[0071] like Figure 2 and Figure 6 As shown, this embodiment provides a joint early warning method, including the following steps:

[0072] S101. According to a preset time threshold and a set collection time point, a plurality of cloud / ground flash monitoring data collected by at least one lightning locator in a monitoring area and a plurality of electric field intensity change data collected by at least one atmospheric electric field meter are obtained.

[0073] Specifically, within the monitoring area, each lightning locator collects and sends multiple cloud / ground flash monitoring data in sequence according to the preset time threshold and the collection time point within the preset time threshold, and each atmospheric electric field meter collects and sends multiple electric field strength change data in sequence according to the preset time threshold and the collection time point within the preset time threshold.

[0074] Furthermore, the cloud / ground lightning monitoring data collected by the lightning locator and the electric field strength change data collected by the atmospheric electric field meter are sent to the generated lightning monitoring database.

[0075] Exemplarily, within the monitoring area, the preset time threshold for data collection by the lightning locator and the atmospheric electric field meter is 20 minutes, and the collection time points are set at intervals of 5 minutes.

[0076] Based on the above-set time parameters, the lightning locator collects and sends cloud / ground flash monitoring data within the range of ≥150km in the five monitoring target warning areas in turn. Among them, when a thunderstorm occurs in the monitoring area within the above-set time parameters, the cloud / ground flash monitoring data at least includes the longitude and latitude coordinates of the cloud / ground flash (i.e., thunderstorm occurrence data). Based on the above-set time parameters, the atmospheric electric field instrument collects and sends the electric field strength change data within the coordinate range of ≤20km of the five monitoring devices in turn.

[0077] S102: Obtain a thunderstorm cloud range map according to thunderstorm coordinate points in each cloud / ground lightning positioning monitoring data.

[0078] Specifically, step S102 includes the following sub-steps:

[0079] S1021. Extract and analyze the cloud / ground lightning monitoring data in the lightning monitoring database in real time, and calculate the Euclidean distance between cloud and ground lightning based on thunderstorm occurrence data.

[0080] Specifically, according to the order of set time thresholds, the cloud / ground flash monitoring data stored in the lightning monitoring database is extracted and analyzed. If the cloud / ground flash monitoring data contains thunderstorm occurrence data (i.e., thunderstorm longitude and latitude coordinate points), it is determined that the cloud / ground flash monitoring data contains thunderstorm occurrence data.

[0081] The Euclidean distance between cloud and ground flashes is calculated using the following formula:

[0082]

[0083] S1022. Perform curve fitting on multiple cloud / ground lightning distances to obtain a fitting curve.

[0084] Specifically, the cloud / ground lightning distance distribution is obtained, and the distance distribution is fitted to obtain a fitting curve.

[0085] S1023. In the fitting curve, take intervals of equal length forward and backward for each point, calculate the variance, and take the distribution variance to obtain the inflection point of the distance curve.

[0086] S1024. Automatically estimate the input parameters required for the clustering model according to the inflection point of the distance curve and the number of dimensions.

[0087] S1025. Perform density clustering according to the input parameters of the clustering model to obtain the thunderstorm cloud structure corresponding to the cloud / ground lightning monitoring data.

[0088] Specifically, the DBSCAN clustering method is used to perform density clustering on the cloud / ground lightning positioning monitoring data to identify the thunderstorm cloud structure corresponding to each cloud / ground lightning positioning monitoring data.

[0089] S1026. Use confidence ellipse to divide the thunderstorm cloud structure to obtain the thunderstorm cloud range map. Specifically, construct a covariance matrix for each cluster point in the thunderstorm cloud cluster, and calculate the eigenvectors in the major axis and minor axis directions of the ellipse according to the covariance matrix. Set half of the major axis of the confidence ellipse to a and half of the minor axis to b. The confidence level can be 95% and the degree of freedom is 2. It can be calculated by chi-square distribution, that is, query the chi-square table significance level equal to 0.05, and there is:

[0090] P(5.991)=0.05

[0091] The ellipse formula is as follows:

[0092] The eigenvector of the major axis of the confidence ellipse is set to v1, and the eigenvalue is λ1; the eigenvector of the minor axis of the confidence ellipse is set to v2, and the eigenvalue is λ2. The major axis is calculated as:

[0093]

[0094] The minor axis of the confidence ellipse is:

[0095]

[0096] Define the angle between the major axis of the confidence ellipse and the positive direction of the x-axis as α. According to trigonometric functions, we have:

[0097]

[0098] The center, major axis, minor axis and confidence region of the confidence ellipse can be determined.

[0099] After determining the center, major axis, minor axis and confidence area of ​​the confidence ellipse, the thunderstorm cloud structure can be placed in the confidence ellipse to obtain a thunderstorm cloud range map.

[0100] The Quickhull algorithm can also be used to obtain the convex hull structure of the thunderstorm area (i.e., the thunderstorm cloud range map). The general principle is: find the two points in the area that are farthest apart. These two points must be part of the convex hull. Then, these two points divide the area into two parts. For each part, select a point in the point set and calculate the distance from this point to a straight line. Then, find the point with the largest distance, which must also be on the convex hull. Repeat the above steps for the area containing this point. Each time, find a new point among the remaining points and split the area until all points are processed. Merge the convex hulls of each part into a whole convex hull.

[0101] S103: Obtaining a thunderstorm cloud trajectory connectivity solution between thunderstorm cloud areas according to a plurality of cloud / ground lightning monitoring data within a preset time threshold.

[0102] Specifically, step S103 includes the following sub-steps:

[0103] S1031. Classify the timing status of thunderstorm occurrence progressively according to the collection timing of each cloud / ground lightning monitoring data.

[0104] Specifically, an empty path state sequence head is constructed, deleted, and added according to the progressive classification result, and the path state sequence is cyclically expanded with a time interval of 5 minutes.

[0105] Furthermore, the received cloud / ground flash monitoring data are parsed according to their order, and if the cloud / ground flash monitoring data contain thunderstorm occurrence data, it is determined whether the lightning locator has collected continuous thunderstorm occurrence data within a preset time threshold.

[0106] The continuous thunderstorm occurrence data is defined as at least two adjacent cloud / ground lightning monitoring data both containing thunderstorm occurrence data, which can be classified into the same progressive classification.

[0107] In the first example, the lightning locator collects a total of 5 cloud / ground flash monitoring data within the current set time parameters during the thunderstorm process, wherein each cloud / ground flash monitoring data contains thunderstorm occurrence data. At this time, it indicates that the lightning locator collects continuous thunderstorm occurrence data.

[0108] In the second example, the lightning locator collects a total of 5 cloud / ground flash monitoring data within the current set time parameters during the thunderstorm process, wherein the first cloud / ground flash monitoring data and the second cloud / ground flash monitoring data contain thunderstorm occurrence data, the third cloud / ground flash monitoring data does not contain thunderstorm occurrence data, and the fourth cloud / ground flash monitoring data and the fifth cloud / ground flash monitoring data contain thunderstorm occurrence data. It can be seen that the adjacent first acquisition time point and the second acquisition time point both have thunderstorms, the first cloud / ground flash monitoring data and the second cloud / ground flash monitoring data contain continuous thunderstorm occurrence data, and the first cloud / ground flash monitoring data and the second cloud / ground flash monitoring data can be divided into the same progressive classification. The adjacent fourth acquisition time point and the fifth acquisition time point both have thunderstorms, the fourth cloud / ground flash monitoring data and the fifth cloud / ground flash monitoring data contain continuous thunderstorm occurrence data, and the fourth cloud / ground flash monitoring data and the fifth cloud / ground flash monitoring data can be divided into the same progressive classification.

[0109] S1032. Calculate the fitness of any two thunderstorm cloud range maps at adjacent time periods.

[0110] Specifically, calculate the fitness S of any two thunderstorm cloud range maps between two adjacent time periods ij .

[0111] Preferably, the fitness is calculated by the difference between the center of mass and the area of ​​the thunderstorm cloud calculated by the Chebyshev distance formula, as follows:

[0112] S ij =w d D d +w s D s

[0113] D d =max([x 1i- x 2j ,y 1i -y 2j )|)

[0114] D s =A 1i -A 2j .

[0115] Where w is the weight coefficient, x is the horizontal coordinate, y is the vertical coordinate, and A is the area of ​​the thunderstorm cloud range map.

[0116] Among them, when the thunderstorm cloud range map is an ellipse, A is the area of ​​the ellipse; when the thunderstorm cloud range map is a convex hull, A is the area of ​​the convex hull.

[0117] S1033. Construct a fitness matrix according to the obtained multiple fitnesses.

[0118] Specifically, after obtaining n1*n2 S ij Then, the fitness matrix is ​​constructed.

[0119] S1034. Use the Viterbi dynamic programming algorithm or the Hungarian matching algorithm to obtain a thunderstorm cloud trajectory connectivity solution between thunderstorm cloud areas.

[0120] In one embodiment, a thunderstorm cloud track connection scheme between thunderstorm cloud regions is obtained according to the Viterbi dynamic programming algorithm, and the general steps include:

[0121] Step 1), initialization, has been completed through progressive classification;

[0122] δ1(j)=w·F1(y0=start, y1=j,x), j=1,2,...,m.

[0123] Step 2), recursively, starting from the second time step, calculate the maximum probability of each state, which depends on the state probability and fitness of the previous time step, update the values ​​in the table, and record the optimal state path;

[0124]

[0125]

[0126] F i (y i-1 ,y i , x)=(j1(y i-1 ,y i ,x,i),f2(y i-1 ,y i ,x,i),…,f K (y i-1 ,y i , x, i)) T .

[0127] Step 3), terminate, find the most likely state in the last time step, and determine the final optimal path;

[0128]

[0129]

[0130] Step 4) Path backtracking: Starting from the state determined by the termination step, backtrack through the path record table to find the most likely state sequence in the entire observation sequence.

[0131]

[0132] In another embodiment, the Hungarian matching algorithm is used to perform multi-path optimal matching of adjacent thunderstorm clouds, and the general steps are:

[0133] Step 1), subtract the minimum value of the row

[0134] For each row, find the lowest element, then subtract it from every element in that row.

[0135] Step 2), subtract the minimum value of the column

[0136] Likewise, for each column, find the lowest element, then subtract it from every element in that column.

[0137] Step 3) Cover all zeros with the minimum number of rows

[0138] Cover all zeros in the resulting matrix using the minimum number of horizontal and vertical lines. If n rows are needed, there is an optimal distribution among the zeros. The algorithm stops.

[0139] If there are less than n rows, go to step 4).

[0140] Step 4), generate extra zeros

[0141] In step 3), find the smallest element that is not covered by a line (this smallest element is defined as k). Subtract the smallest element k from all the undiscovered elements, and then add the smallest element k to all the elements that are covered twice.

[0142] In the second example, through the above steps S1033 and S1034, it can be known that the thunderstorm clouds at the first collection time point and the second collection time point have a matching relationship, which can form a first thunderstorm cloud path, and the thunderstorm clouds at the fourth collection time point and the fifth collection time point have a matching relationship, which can form a second thunderstorm cloud path.

[0143] In addition, since the third collection time point does not include thunderstorm occurrence data, but since the second collection time point and the fourth collection time point include thunderstorm occurrence data, a virtual thunderstorm cloud can be set between the second collection time point and the fourth collection time point, so that a complete thunderstorm cloud path including the first thunderstorm cloud path, a virtual thunderstorm cloud and the second thunderstorm cloud path is formed within the current preset time threshold during the thunderstorm occurrence process.

[0144] S104. Construct a thunderstorm cloud motion trajectory extrapolation model according to the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas.

[0145] Specifically, the decaying trend exponential smoothing method is used to construct the trajectory extrapolation model of the current multi-path thunderstorm cloud. The thunderstorm trajectory extrapolation adopts the decaying trend exponential smoothing method, which is an improvement of the exponential smoothing method. When the exponential smoothing method is used for prediction, it is assumed that the prediction trend is always maintained. However, when the prediction period is too long, it is obvious that the trend will weaken as the prediction period increases, and the model prediction accuracy will also decrease. Therefore, the decaying trend exponential smoothing method will gradually weaken the influence of the long-term trend, which will help improve the prediction results.

[0146] The following is its common form, the formula for Holt's linear smoothing method:

[0147] Horizontal smoothing formula:

[0148] y t+1 =αy t +(1-α)(y t +b t )

[0149] Among them, y t+1 is the predicted value at time t, α is the horizontal smoothing coefficient, y t is the actual observed value, b t is the trend component.

[0150] Trend smoothing formula:

[0151] b t+1 =β(y t+1 -yt )+(1-β)b t

[0152] Among them, β trend smoothing coefficient, b t is the trend component at time t.

[0153] Prediction formula, predicting the value at m time points in the future:

[0154] y t+m =y t +m×b t .

[0155] In addition, step S102 and step S103 can be repeated according to multiple cloud / ground flash monitoring data collected at the set collection time point within the preset time threshold, so as to obtain a thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas according to the multiple cloud / ground flash monitoring data collected at the set collection time point, and the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas is input into the thunderstorm cloud motion trajectory extrapolation model of step S104, thereby updating the dynamic prediction information of the thunderstorm cloud tracking trajectory in the thunderstorm cloud motion trajectory extrapolation model.

[0156] Exemplarily, the collection time points are set at intervals of 5 minutes within the preset time threshold. Therefore, step S102 and step S103 are repeated every 5 minutes, and the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas is obtained based on the cloud / ground flash monitoring data sent by the received lightning monitoring database. The thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas is input into the thunderstorm cloud motion trajectory extrapolation model of step S104, thereby updating the dynamic prediction information of the thunderstorm cloud tracking trajectory in the thunderstorm cloud motion trajectory extrapolation model.

[0157] S105. Eliminate data residuals in the electric field intensity change data and construct an electric field early warning model.

[0158] Specifically, the integrated empirical mode decomposition method is used to calculate and obtain the data residuals of the atmospheric electric field instrument in the electric field strength change data caused by environmental factors, which is convenient for eliminating the data residuals in the electric field strength change data and effectively solving the impact of the environment on the atmospheric electric field instrument, so as to obtain more accurate electric field strength change data.

[0159] Furthermore, the integrated empirical mode decomposition method is the following two repeated steps:

[0160] Step 1) Find all the local maximum and minimum values ​​in the original signal, and construct the upper envelope and the lower envelope by combining the extreme points through the curve fitting method, so that the original signal is enveloped by the upper and lower envelopes;

[0161] Step 2) The upper and lower envelopes can be used to construct their mean curves, and the mean curve is subtracted from the original signal. The resulting curve is the IMF.

[0162] In the integrated empirical mode decomposition method, the ability to obtain a reasonable IMF depends on the distribution of the signal extreme points. If the signal extreme points are unevenly distributed, mode aliasing will occur. The integrated empirical mode decomposition method adds white noise to the signal to be decomposed and uses the uniform distribution of the white noise spectrum. When the signal is added to the white noise background that is uniformly distributed throughout the time-frequency space, the signals of different time scales will automatically be distributed to the appropriate reference scale, and due to the characteristics of zero-mean noise, after multiple averaging, the noise will cancel each other out, and the result of the integrated mean can be used as the final result.

[0163] S106. Obtaining an electric field amplitude according to the electric field strength change data, and generating electric field amplitude warning information when the electric field amplitude meets a preset condition.

[0164] Specifically, step S106 includes the following sub-steps:

[0165] S1061, obtaining each electric field amplitude corresponding to each electric field intensity change data within a preset time threshold;

[0166] S1062, comparing each electric field amplitude with a preset electric field amplitude threshold value to obtain a determination quantity value of the electric field amplitude greater than the preset electric field amplitude threshold value;

[0167] S1063, obtaining a ratio of the above determination quantity value to all electric field amplitudes within a preset time threshold;

[0168] S1064: Compare the above-mentioned proportional value with a preset proportional value threshold, and when the above-mentioned proportional value is greater than the preset proportional value threshold, generate electric field amplitude warning information.

[0169] Exemplarily, within the monitoring area, the collection time points are set to be every 5 minutes, the collection time period is 20 minutes, and a total of 5 electric field intensity change data are obtained.

[0170] By analyzing the 5 electric field intensity change data, 5 electric field amplitudes are obtained;

[0171] The preset electric field amplitude threshold is set to 0.7 kV / m, and the above five electric field amplitudes are respectively compared with the preset electric field amplitude threshold, and the determination value of the electric field amplitude greater than the preset electric field amplitude threshold is obtained to be 3;

[0172] The ratio of the above determination quantity value to all electric field amplitudes within the preset time threshold is obtained to be 3 / 5;

[0173] The preset proportional numerical threshold is set to 1 / 3, the above proportional numerical value 3 / 5 is compared with the preset proportional numerical threshold 1 / 3, and it is determined that the above proportional numerical value 3 / 5 is greater than the preset proportional numerical threshold 1 / 3, thereby generating electric field amplitude warning information.

[0174] In this embodiment, multiple comparisons are performed on multiple electric field amplitudes to improve the accuracy of the electric field amplitude warning. In addition, the warning retention time can be increased to determine whether the warning events can be merged to maintain consistency.

[0175] In addition, in order to further improve the accuracy of the electric field amplitude warning, at least one electric field amplitude in the previous preset time threshold adjacent to the current preset time threshold, and multiple electric field amplitudes in the current preset time threshold can be selected and compared with the preset differential value threshold respectively.

[0176] S107, obtaining an electric field differential value, and generating electric field differential warning information when the electric field differential value meets a preset condition.

[0177] Specifically, step S107 includes the following sub-steps:

[0178] S1071, using the following formula, obtaining each electric field differential value according to each electric field amplitude obtained above;

[0179]

[0180] In the above formula, E is the electric field value at any time, and t is the time.

[0181] S1072, comparing each electric field difference value with each preset electric field difference value threshold value, to obtain a determination quantity value of the electric field difference value greater than the preset electric field difference value threshold value;

[0182] S1073, obtaining a ratio of the above determination quantity value to all electric field difference values ​​within a preset time threshold;

[0183] S1074. Compare the above-mentioned proportional value with a preset proportional value threshold value, and when the above-mentioned proportional value is greater than the preset proportional value threshold value, generate electric field differential value warning information.

[0184] Exemplarily, based on the example of step S106, a total of 5 electric field amplitudes are obtained.

[0185] Using the above formula, a total of 5 electric field difference values ​​are obtained;

[0186] The preset electric field amplitude threshold is set to 0.12 kV / (m*s), and the above five electric field differential values ​​are compared with the preset electric field differential value threshold respectively, and the determination value of the number of electric field differential values ​​greater than the preset electric field differential value threshold is obtained to be 3;

[0187] The ratio of the above determination quantity value to all electric field difference values ​​within the preset time threshold is obtained to be 3 / 5;

[0188] The preset proportional numerical threshold is set to 1 / 4, the above proportional numerical value 3 / 5 is compared with the preset proportional numerical threshold 1 / 4, and it is determined that the above proportional numerical value 3 / 5 is greater than the preset proportional numerical threshold 1 / 4, thereby generating electric field amplitude warning information.

[0189] In this embodiment, multiple comparisons are performed on multiple electric field differential values ​​to improve the accuracy of the electric field differential value warning. In addition, the warning retention time can be increased to determine whether the warning events can be merged to maintain consistency.

[0190] In addition, in order to further improve the accuracy of the electric field differential value warning, at least one electric field differential value in the previous preset time threshold adjacent to the current preset time threshold can be selected, and multiple electric field differential values ​​in the current preset time threshold can be compared with the preset differential value threshold respectively.

[0191] S108. The joint warning general model performs a joint warning based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information.

[0192] S109. Display the geographical information of the monitoring area, the thunderstorm cloud tracking trajectory, the electric field amplitude and the electric field differential value.

[0193] Due to the randomness of the shape and charge distribution of thunderstorm clouds, it is difficult to establish a linear relationship using the electric field value to estimate the distance to the thunderstorm. The electric field warning model does not grade the electric field, but increases the alarm retention time to obtain the continuity of the electric field warning, which can further solve the problem of unstable warning caused by the influence of the environment on the electric field.

[0194] In this embodiment, since the dynamic prediction information of the thunderstorm cloud tracking trajectory can be set according to the preset time threshold, the joint warning general model can be combined according to the thunderstorm cloud tracking trajectory within the preset time threshold, and the electric field amplitude warning results and electric field differential warning results obtained within the preset time threshold to obtain a more accurate lightning warning result.

[0195] In addition, in the first embodiment, before the implementation of step S108, after obtaining a priori values ​​of multiple historical data of the monitoring area according to the verification index, a joint early warning general model is constructed, including the following steps:

[0196] The thunderstorm occurrence data of several historical thunderstorm events in the monitoring area are obtained, and each thunderstorm occurrence data is defined as verification data;

[0197] Each verification data is verified according to the verification index to obtain each priori value. Among them, the verification index at least includes thunderstorm hit rate, thunderstorm false alarm rate and thunderstorm false alarm rate, and each priori value includes the corresponding thunderstorm hit rate value, thunderstorm false alarm rate value and thunderstorm false alarm rate value;

[0198] A general joint early warning model is constructed based on the several prior values ​​obtained above.

[0199] Since there are several historical thunderstorm events, there are also several prior values ​​obtained based on the thunderstorm occurrence data of the historical thunderstorm events and after verification of the thunderstorm occurrence data. The corresponding joint warning general model includes several prior values.

[0200] Exemplarily, one thunderstorm occurrence data of a historical thunderstorm event in the monitoring area is extracted and used as verification data. The prior value obtained after the verification data is verified according to the verification index is set to 2, wherein the thunderstorm hit rate value corresponding to the prior value is 83%, the thunderstorm false alarm rate value is 17%, and the thunderstorm false alarm rate value is 13%.

[0201] In addition, in the above-mentioned first embodiment, iteratively updating the joint warning general model according to the comparison result of the cloud / ground lightning monitoring data includes the following steps:

[0202] Step 1) multiple cloud / ground lightning monitoring data obtain current verification values ​​according to the verification indicators.

[0203] Specifically, each cloud / ground lightning monitoring data obtained within a preset time threshold is used as verification data, and each verification data is verified according to the above-mentioned verification indicators to obtain each verification value.

[0204] Among them, each verification value includes the corresponding thunderstorm hit rate value, thunderstorm misalarm rate value and thunderstorm false alarm rate value.

[0205] Exemplarily, based on the above-mentioned step S103, within the preset time threshold, a total of 4 cloud / ground flash monitoring data including thunderstorm occurrence data are obtained, and 4 verification values ​​are obtained after the 4 cloud / ground flash monitoring data are verified according to the verification indicators, and each verification value includes the corresponding thunderstorm hit rate value, thunderstorm false alarm rate value and thunderstorm false alarm rate value.

[0206] Step 2) The current verification value is compared with the selected verification value and the neighboring values ​​of the selected verification value, and the joint warning general model is iteratively updated according to the comparison results.

[0207] Specifically, step 2) includes the following sub-steps:

[0208] Step 21), select a first verification value in the joint early warning general model according to the location of the monitoring area, and select at least one adjacent value before and after the selected first verification value (ie, the first verification value and the second verification value);

[0209] Exemplarily, in one of the embodiments, when selecting a selected prior value, first, the average value of several prior values ​​in the joint early warning model is obtained, and then, if it is determined that several prior values ​​in the joint early warning model include a prior value corresponding to the average value, it is used as the selected prior verification value; otherwise, among several prior values ​​in the joint early warning model, the prior value closest to the average value is selected as the selected prior verification value.

[0210] Exemplarily, in another embodiment, when selecting the selected priori value, since the priori value has been saved in the priori value list of the joint early warning model, a priori value can be randomly selected from the priori value list as the selected priori value.

[0211] Step 22), obtaining the average value of the sum of the thunderstorm hit rate value, the thunderstorm false alarm rate value and the thunderstorm false alarm rate value in each current verification value, and defining it as the current average value;

[0212] Step 23), respectively obtain the average values ​​of the sum of the thunderstorm hit rate value, the thunderstorm misalarm rate value and the thunderstorm false alarm rate value in the selected first verification value, the front first verification value and the rear first verification value, and define them as the selected average value, the front selected average value and the rear selected average value respectively;

[0213] Step 24), compare each current average value with the selected average value, the previous selected average value and the subsequent selected average value respectively. When a current average value is greater than the selected average value, the previous selected average value and the subsequent selected average value, at least the current verification value corresponding to the current average value is replaced with the selected first verification value, and the current joint early warning general model is iteratively updated until convergence.

[0214] In the above step 2), if each current average value is smaller than the selected average value, the previous selected average value and the next selected average value, the selected average value is not replaced, and the current joint early warning general model is not iteratively updated.

[0215] When iteratively updating the joint early warning general model, illustratively, each current average value is compared with the selected average value, the previous selected average value and the subsequent selected average value, respectively. If the current average value is greater than the selected average value, the previous selected average value and the subsequent selected average value by 1%, at least the current verification value corresponding to the current average value is replaced by the selected first verification value, and the current joint early warning general model is iteratively updated until convergence.

[0216] In the first embodiment, since the thunderstorm cloud motion trajectory extrapolation model can receive cloud / ground flash monitoring data collected according to a preset time threshold, the thunderstorm cloud motion trajectory extrapolation model can iteratively update the dynamic prediction information of the thunderstorm cloud tracking trajectory formed according to the cloud / ground flash monitoring data. At the same time, after receiving the cloud / ground flash monitoring data collected within the preset time threshold, the joint warning general model can also iteratively update the current joint warning general model according to the comparison result.

[0217] By constructing a joint warning general model and iteratively updating the joint warning general model, the multi-parameter numerical problem of being difficult to effectively obtain complex models in certain areas (for example, the northern region) where there are few thunderstorm events can be solved. In the above method of constructing and iteratively updating the joint warning general model, the input of parameters and training are minimized as much as possible (such as the parameters of the thunderstorm identification algorithm and the extrapolation algorithm in the trajectory extrapolation algorithm are obtained by automatic estimation, and the electric field warning cancels the graded warning, etc.), and a general model is established based on thunderstorm data from other regions. Then, while the business is running normally, the model parameters are iteratively updated under different regional terrains until convergence, thereby obtaining a joint warning model suitable for the region.

[0218] like Figure 3 As shown, the present invention provides a joint early warning device, comprising:

[0219] A lightning locator 1 is used to collect and send multiple cloud / ground lightning monitoring data in the monitoring area according to a preset time threshold;

[0220] Atmospheric electric field instrument 2, used to collect and send multiple electric field intensity change data in the monitoring area according to a preset time threshold;

[0221] The collection and distribution module 3 is used to receive, save and distribute a plurality of cloud / ground lightning monitoring data sent by the lightning locator and a plurality of electric field strength change data sent by the atmospheric electric field meter.

[0222] The thunderstorm cloud motion trajectory extrapolation module 4 is used to construct a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory after obtaining the thunderstorm cloud trajectory connection scheme between thunderstorm cloud areas according to each cloud / ground lightning monitoring data;

[0223] The electric field warning module 5 is used to eliminate the data residual in each electric field strength change data, build an electric field warning model, and generate electric field warning information including at least the electric field amplitude and the electric field difference value according to preset conditions;

[0224] A joint warning module 6 is used to conduct a joint warning based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information;

[0225] Display module 7 is used to display the geographical information of the monitoring area, the thunderstorm cloud tracking trajectory, the electric field amplitude and the electric field differential value.

[0226] Embodiment 2

[0227] like Figure 4 and Figure 6 As shown, this embodiment provides a joint warning device, which is used to implement the joint warning method in the above embodiment 1, including:

[0228] The lightning locator 1 collects and sends a plurality of cloud / ground lightning monitoring data detected in the monitoring area according to a preset time threshold and a set collection time point.

[0229] The atmospheric electric field instrument 2 collects and sends a plurality of electric field intensity change data monitored in the monitoring area according to a preset time threshold and a set collection time point.

[0230] The collection and distribution module 3 is used to receive, save and distribute a plurality of cloud / ground lightning monitoring data sent by the lightning locator and a plurality of electric field strength change data sent by the atmospheric electric field meter.

[0231] The collection and distribution module 3 includes a lightning monitoring database for storing cloud / ground lightning monitoring data collected and sent by a lightning locator, and electric field strength change data collected and sent by an atmospheric electric field meter.

[0232] The thunderstorm cloud motion trajectory extrapolation module is used to obtain the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on the cloud / ground lightning monitoring data, and then construct a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory.

[0233] Specifically, the thunderstorm cloud motion trajectory extrapolation module includes a thunderstorm cloud range map acquisition module 41, a thunderstorm cloud trajectory connectivity scheme acquisition module 42 and a thunderstorm cloud motion trajectory extrapolation model construction module 43.

[0234] Furthermore, the thunderstorm cloud range map obtaining module 41 is used to implement the implementation steps of obtaining the thunderstorm cloud range map according to the thunderstorm coordinate points in the cloud / ground lightning positioning monitoring data recorded in step S102 in the first embodiment.

[0235] The thunderstorm cloud trajectory connectivity scheme acquisition module 42 is used to implement the implementation steps recorded in step S103 of the above-mentioned embodiment 1 of obtaining the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas according to multiple cloud / ground lightning monitoring data within a preset time threshold.

[0236] The thunderstorm cloud motion trajectory extrapolation model construction module 43 is used to implement the implementation steps of constructing the thunderstorm cloud motion trajectory extrapolation model according to the thunderstorm cloud trajectory connection scheme between thunderstorm cloud areas recorded in step S104 in the above-mentioned embodiment 1.

[0237] The electric field warning module is used to eliminate data residuals in the electric field strength change data and generate electric field warning information including at least the electric field amplitude and the electric field differential value according to preset conditions.

[0238] Specifically, the electric field warning module includes a data residual elimination module 51 , an electric field amplitude warning information generation module 52 and an electric field difference warning information generation module 53 .

[0239] Furthermore, the data residual elimination module 51 is used to implement the implementation steps of eliminating the data residual in the electric field intensity variation data and constructing the electric field early warning model as recorded in step S105 in the first embodiment.

[0240] The electric field amplitude warning information generating module 52 is used to implement the step of obtaining the electric field amplitude according to the electric field strength change data as recorded in step S106 in the first embodiment above, and generating the electric field amplitude warning information when the electric field amplitude meets the preset conditions.

[0241] The electric field differential warning information generating module 53 is used to implement the step of obtaining the electric field differential value recorded in step S107 in the first embodiment, and when the electric field differential value meets the preset conditions, generating the electric field differential warning information.

[0242] The joint warning module 6 is used to make a joint warning based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information.

[0243] Display module 7 is used to display the geographical information of the monitoring area, the thunderstorm cloud tracking trajectory, the electric field amplitude and the electric field differential value.

[0244] In addition, in the second embodiment, the joint warning module 6 also includes a joint warning general model construction module and a joint warning general model iterative update module. Among them, the joint warning general model construction module implements the implementation steps recorded in the above embodiment one to construct the joint warning general model after obtaining a priori values ​​according to the verification indicators based on multiple historical data of the monitoring area. The joint warning general model iterative update module implements the implementation steps recorded in the above embodiment one to iteratively update the joint warning general model according to the comparison results of cloud / ground lightning monitoring data.

[0245] like Figure 5 As shown, the present embodiment discloses an electronic device, which includes a CPU 8, which can perform various appropriate actions and processes according to the program stored in the ROM 9 or the program loaded from the storage part 14 into the RAM 10. Various programs and data required for system operation are also stored in the RAM 10. The CPU 8, the ROM 9 and the RAM 10 are connected to each other through a bus. The I / O interface 11 is also connected to the bus.

[0246] The following components are connected to the I / O interface 11, which include an input section 12 such as a keyboard, a mouse, etc.; an output section 13 including such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 14 including a hard disk, etc.; and a communication section 15 including a network interface card such as a local area network (LAN) card, a modem, etc. The communication section 15 performs communication processing via a network such as the Internet. A drive 16 is also connected to the I / O interface 11 as needed. A removable medium 17, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 16 as needed, so that a computer program read therefrom is installed into the storage section 14 as needed.

[0247] In particular, according to an embodiment of the present application, the above Figure 1 and Figure 2 The steps of the joint early warning method described in the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 15, and / or installed from the removable medium 17. When the computer program is executed by the CPU 8, the above-mentioned method defined in the apparatus of the present application is executed. Figure 1 and Figure 2 The steps of the joint early warning method are described in.

[0248] This embodiment provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, for example, including a memory storing a computer program, and the computer program can be executed by a processor to complete the steps of the aforementioned method. The computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash, Memory, magnetic surface storage, optical disk, CD-ROM, etc.

[0249] Those skilled in the art will appreciate that all or part of the steps of the above method embodiments may be implemented by hardware associated with program instructions, and the above program may be stored in a computer-readable storage medium. Figure 1 and Figure 2 The steps of the joint early warning method recorded in; and the aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, disks or optical disks.

[0250] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for an electronic device (which can be a personal computer, server, network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, disks or optical disks.

[0251] The above description is only a preferred embodiment of the present invention, which is only illustrative and not restrictive of the present invention. Those skilled in the art understand that many changes, modifications, and even equivalences may be made to the present invention within the spirit and scope defined by the claims of the present invention, but all of them will fall within the scope of protection of the present invention.

Claims

1. A joint early warning method, characterized in that: The following steps are involved: According to a preset time threshold, a plurality of cloud / ground lightning monitoring data collected by at least one lightning locator in the monitoring area and a plurality of electric field intensity change data collected by at least one atmospheric electric field meter are obtained; After obtaining the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on the cloud / ground lightning monitoring data, a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory is constructed; After eliminating the data residuals in each electric field intensity change data, an electric field early warning model is constructed, and electric field early warning information including at least an electric field amplitude and an electric field differential value is generated according to preset conditions; A joint warning is issued based on the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information.

2. The joint early warning method according to claim 1, characterized in that: In the scheme for obtaining thunderstorm cloud trajectory connectivity between thunderstorm cloud areas according to each cloud / ground lightning monitoring data, it includes: obtaining a thunderstorm cloud range map according to thunderstorm coordinate points in each cloud / ground lightning positioning monitoring data, including: Obtain the distance value between two adjacent thunderstorm coordinate points in each cloud / ground lightning location monitoring data; Perform curve fitting on multiple distance values ​​to obtain a fitting curve; In the fitting curve, take the intervals of equal length forward and backward from each point, and calculate the variance, and take the distribution variance to obtain the inflection point of the distance curve; The input parameters of the clustering model are obtained by calculating the inflection point of the distance curve; Density clustering is performed according to the input parameters of the clustering model to obtain the thunderstorm cloud structure corresponding to the cloud / ground lightning monitoring data; The thunderstorm cloud structure is divided using the confidence ellipse or Quickhull algorithm to obtain the thunderstorm cloud range map.

3. The joint early warning method according to claim 2, characterized in that: The scheme for connecting thunderstorm cloud trajectories between thunderstorm cloud areas according to the cloud / ground lightning monitoring data also includes: The thunderstorm occurrence timing state is progressively classified according to the collection timing of each cloud / ground lightning monitoring data; Calculate the fitness of any two thunderstorm cloud range maps at adjacent times; A fitness matrix is ​​constructed based on the multiple fitness measures obtained; The Viterbi dynamic programming algorithm or the Hungarian matching algorithm is used to obtain the thunderstorm cloud track connectivity scheme between thunderstorm cloud areas.

4. The joint early warning method according to claim 1, characterized in that: In said generating electric field warning information including at least electric field amplitude and electric field differential value according to preset conditions, including: The electric field amplitude is obtained according to the electric field strength change data, and when the electric field amplitude meets the preset conditions, the electric field amplitude warning information is generated; An electric field differential value is obtained according to the electric field amplitude, and when the electric field differential value meets a preset condition, an electric field differential value warning information is generated.

5. The joint early warning method according to claim 4, characterized in that: In the case where the electric field amplitude meets the preset condition, it includes: At least comparing the plurality of electric field amplitudes within a preset time threshold with preset amplitude thresholds respectively, and generating electric field amplitude warning information when the preset conditions are met; In the case where the electric field difference value meets a preset condition, it includes: At least comparing the plurality of electric field differential values ​​within a preset time threshold with preset differential value thresholds, and generating electric field amplitude warning information when a preset condition is met; The plurality of prior electric field difference values ​​correspond to the plurality of prior electric field amplitudes.

6. The joint early warning method according to claim 1, characterized in that: In the joint warning according to the dynamic prediction information of the thunderstorm cloud tracking trajectory and the electric field warning information, it also includes: After obtaining prior values ​​of multiple historical data in the monitoring area according to the verification indicators, a joint early warning general model is constructed; The cloud / ground lightning monitoring data within the preset time threshold obtains the current verification value according to the verification index; The current verification value is compared with the prior value and the adjacent value of the selected prior value respectively. If the current verification value is greater than the prior value and the adjacent value of the prior value, the current verification value is used as the prior value to iteratively update the joint warning general model.

7. The joint early warning method according to claim 6, characterized in that: The verification index at least includes verifying the hit rate, false alarm rate and false alarm rate of the historical data and the cloud / ground lightning monitoring data; The current verification value, the priori value and the neighboring values ​​of the priori value at least include a hit rate value, an error rate value and a false alarm rate value.

8. The joint early warning method according to claim 1, characterized in that: Also includes: Displays the geographical information of the monitoring area, thunderstorm cloud tracking trajectory, electric field amplitude and electric field differential value.

9. A device for implementing the joint early warning method according to any one of claims 1 to 8, characterized in that: include: Lightning locator, which collects and sends multiple cloud / ground lightning monitoring data in the monitoring area according to the preset time threshold; The atmospheric electric field instrument collects and sends multiple electric field intensity change data within the monitoring area according to the preset time threshold; The thunderstorm cloud motion trajectory extrapolation module is used to obtain the thunderstorm cloud trajectory connectivity scheme between thunderstorm cloud areas based on the cloud / ground lightning monitoring data, and then construct a thunderstorm cloud motion trajectory extrapolation model containing dynamic prediction information of thunderstorm cloud tracking trajectory; The electric field warning module is used to eliminate the data residuals in the electric field strength change data, build an electric field warning model, and generate electric field warning information including at least the electric field amplitude and the electric field differential value according to preset conditions; The joint warning module is used to provide joint warning based on the dynamic prediction information of thunderstorm cloud tracking trajectory and electric field warning information.

10. The device according to claim 9, characterized in that Also includes: The display module is used to display the geographic information of the monitoring area, thunderstorm cloud tracking trajectory, electric field amplitude and electric field differential value.

11. An electronic device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the joint early warning method as claimed in any one of claims 1 to 8.

12. A storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the joint early warning method as claimed in any one of claims 1 to 8.

Images