98 results about "Thunderstorm" patented technology

The invention discloses a thunderstorm and gale dynamic extrapolation forecasting method based on multi-task MSTA-ConvLSTM, and relates to the technical field of atmospheric sciences, and the method comprises the steps: constructing a multi-source fusion spatio-temporal data set and a gale process record file, and obtaining a data set; a thunderstorm gale nowcasting model based on an MSTA-ConvLSTM model is constructed; according to the method, a dynamic gating mechanism of lightning data is introduced, whether a lightning probability prediction task is started or not is intelligently judged according to the lightning activity intensity, thunderstorm, gale and co-evolution characteristics of the lightning process are considered, 0-3-hour high-temporal-spatial-resolution nowcasting is achieved, the limitation of traditional single-task forecasting is broken through, a multi-task learning structure is adopted, and the prediction efficiency is improved. The model is guided to recognize the lightning occurrence probability during main task gale prediction, and the overall perception capability of a severe convection system is improved; extreme sample learning is enhanced through a weighted loss function, and output is optimized in combination with a gust coefficient model, so that the forecasting precision and practicability are effectively improved.

The invention discloses a thunderstorm gale regression analysis method and system based on multi-stream parameters, and relates to the technical field of data processing. The method comprises the following steps: collecting multi-stream parameters in real time, and constructing a real-time multi-stream parameter sequence; mapping the real-time multi-flow-parameter sequence to a pre-constructed multi-flow-parameter coupling sensitivity field model, and calculating to generate a real-time convection sensitivity field; inputting the real-time convection sensitivity field into the physical consistency constraint model to generate real-time structured convection characteristics; main regression analysis under multi-flow parameter regression is executed, and a main regression result is established; and after the real-time local disturbance index is calculated, disturbance residual compensation of the main regression result is triggered, dynamic weight adjustment is carried out on the main regression result after residual compensation by utilizing seasonal phase characteristics, and a thunderstorm gale regression analysis result is output. The technical problem that thunderstorm gale prediction accuracy is insufficient in the prior art is solved, and the technical effect of improving thunderstorm gale regression analysis result accuracy is achieved.

The utility model provides a three-core aviation plug connector, which is applied to the field of connectors and comprises a male seat rubber coating, a male seat plastic shell, a female head inner core, a grounding pin, a male seat female pin, a female head male pin, a female head female pin, a locking fastener, a wire clamp and a female head rubber coating. One side of the male seat rubber coating is an input side, the other side of the male seat rubber coating is a butt joint side, the butt joint side is matched with one side of the female head inner core, the male seat plastic shell is arranged at the connecting position of the male seat rubber coating and the female head inner core, the female head inner core is embedded in the locking fastener, and the male seat rubber coating is matched with the female head inner core. The two ends of the grounding pin and the two ends of the male base female pin are connected into the male base rubber coating and the female head inner core respectively, the female head male pin and the female head female pin are connected with the male base female pin and the grounding pin respectively, and one end of the female head male pin and one end of the female head female pin are inserted into the female head inner core. The technical problems that at present, if contact is poor, signals are prone to being lost, the signals can be interfered in thunderstorm weather, and the signals are unstable are solved.

The invention discloses a vehicle-mounted micrometeorological monitoring and camping site disaster early warning system and method, and belongs to the technical field of vehicle-mounted meteorological monitoring, and the system comprises a data collection and preprocessing module which is used for obtaining and calibrating multi-source meteorological and geographic data; the microclimate modeling module is used for generating a microclimate grid with a real-time weather predicted value by taking local microclimate data as a benchmark and fusing cloud and geographic data; the disaster risk assessment module performs quantitative risk assessment on camping disasters such as flood, tree smashing, landslide and thunderstorm based on the data of the microclimate grid; and the early warning generation module generates visual early warning information including a disaster icon, a risk level, time information and a risk avoiding instruction according to the quantitative risk assessment result. According to the invention, by optimizing the layout of the sensors and introducing a terrain correction interpolation algorithm and scene disaster assessment, the meteorological monitoring precision in a mobile environment is improved, and timely early warning of camping field micro-scale disasters is realized.

The utility model belongs to the field of water tanks, particularly relates to a lightning-protection assembly type bolted connection water tank, and aims to solve the problems that the top of an existing traditional water tank is usually provided with a maintenance port which is convenient for personnel to enter the water tank and maintain internal electric appliances, but the personnel usually carry a tool box by hands during maintenance, and the maintenance is inconvenient. In order to solve the problems that when a person climbs to the top of a water tank through a crawling ladder, the person needs to lift a tool box all the time, the tool box is inconvenient to climb, potential safety hazards are likely to occur, and the water tank cannot be subjected to lightning protection and is likely to be damaged in the thunderstorm weather using process, the following scheme is provided that the tool box comprises a foundation platform, and channel steel is fixedly arranged at the top of the foundation platform; a worker can fix the tool box to the tail end of the connecting rope through the transportation mechanism, the tool box does not need to be held by hand in the climbing process, the instability risk caused by holding with one hand is avoided, lightning current can be directly guided into the ground through the lightning protection mechanism, and the water tank and internal equipment are prevented from being damaged by lightning stroke.

The invention relates to the field of large steel structure storage tanks (tanks), and discloses a tank upside-down lifting installation construction process. The method comprises the following steps that S1, inspection before construction is conducted, specifically, when the thickness of a steel plate for the groove wall in main construction materials is larger than or equal to 22 mmm, ultrasonic inspection should be conducted according to NB / T470133, and the quality grade should meet the III-grade requirement; checking the number: when the thickness is less than or equal to 30mm, carrying out spot check on not less than two steel plates according to 20% of the number of the steel plates, and checking the steel plates one by one when disqualification is found; when the thickness is greater than 30mm, checking one by one; the Q345R steel plate with the thickness larger than or equal to 36 mmmm should be used in the normalizing state. Related hoisting calculation is carried out according to a design drawing, a hygrothermograph is additionally arranged in the process and used for monitoring the surrounding environment of construction, the weather conditions of the same season in the first few years are extracted in the early stage of construction, and shutdown materials need to be prepared in advance to protect facilities in rainy seasons and thunderstorm seasons.

The invention discloses a thunderstorm weather assessment method based on rainfall and atmospheric electric field feature weighted fusion. The method comprises the following steps: determining a target area; multi-modal features are extracted; and an IGBDT model is constructed. The invention provides a technical scheme with high physical interpretability and excellent classification performance for severe convection weather monitoring.

The application discloses a hail cloud early identification method based on polarization radar, comprising the following steps: establishing a sample library; extracting multiple determination factors of hail cloud monomer radar data 20 minutes before hail occurrence in the sample library; establishing a logistic regression model for multiple determination factors in the sample library; selecting an optimal weight algorithm by using a cross-validation mechanism; calculating the weight of multiple determination factors; determining a hail early warning index threshold; comparing the early warning index with the hail early warning index threshold; if the early warning index is greater than or equal to the hail early warning index threshold, the early hail cloud monomer is determined; otherwise, the general thunderstorm cloud is determined. The application can effectively improve the early hail cloud identification accuracy, accurately distinguish the hail cloud from the general thunderstorm cloud, and assist weather forecasting and artificial hail prevention operation based on polarization radar data and three key determination factors.

The invention discloses a lightning disaster early warning protection system and method, and the system comprises a lightning parameter collection module which is used for obtaining the atmospheric electric field intensity, the lightning occurrence position and the thunderstorm movement track in real time; the environment parameter monitoring module is used for collecting surrounding environment data of the high-risk place; the data processing center is used for being connected with the thunder and lightning parameter acquisition module and the environmental parameter monitoring module, a thunder and lightning early warning model based on an improved random forest algorithm is built in the data processing center, and grading early warning information is generated by fusing and analyzing thunder and lightning parameters and environmental parameters; the early warning terminal is used for receiving and displaying the graded early warning information and triggering an alarm when the early warning grade reaches a preset threshold value; and the protection execution mechanism is used for executing targeted protection action according to the instruction of the data processing center, and is also used for gradually relieving the protection state according to a preset program after the data processing center detects that the thunder threat is relieved.

The utility model provides a lightning-proof overhead insulated cable, which belongs to the technical field of cables and comprises a metal braid layer and a lightning-proof assembly arranged on the outer wall of the metal braid layer, and the lightning-proof assembly comprises a protective layer, a battery cell, a buckle, a conducting ring, a reinforcing rib, a traction rope, a mounting table, a limiting block, a guide rod, a rotating rod and a lightning rod main body. Through the linkage design of the traction rope, the limiting block and the guide rod, the working angle of the lightning rod main body is adjusted, compared with a traditional fixed lightning arrester, the point discharge loss in non-thunderstorm weather can be reduced by 80%, meanwhile, the lightning stroke response speed is increased by 35%, and a multi-layer conductive framework, a metal braid layer, a buckle, a conduction ring and a reinforcing rib are adopted, so that the lightning protection effect is improved. Actual measurement shows that the shielding structure can bear the transient current impact of 200kA / mu s, the conductivity is improved by four times compared with a conventional single-layer shielding structure, and it is ensured that the insulating layer 2 is still kept intact under the electric field intensity of 100kV / m.

The application discloses a heavy-load railway catenary reliability prediction method and system under thunder weather, and the method comprises the following steps: acquiring historical lightning strike data of the heavy-load railway catenary, and establishing a thunder generation model according to the historical lightning strike data of the heavy-load railway catenary; continuously dividing and sorting time periods of thunder weather at equal intervals, in each time period, using a density-based clustering algorithm to perform spatial clustering on lightning position information to obtain a thunder clustering cluster of each time period; combining thunder parameters to identify a region in which thunder occurs relatively densely in each clustering cluster, and then predicting a lightning range of a next time period through thunder clustering clusters of previous continuous time periods; combining the predicted lightning range and the thunder generation model to calculate a lightning strike failure probability of the heavy-load railway catenary; and calculating a reliability of the heavy-load railway catenary according to the lightning strike failure probability. The method has high usability and accuracy.

The application discloses a thunderstorm prediction method based on a space-time memory decoupling RNN, and the method steps comprise the following steps: acquiring meteorological radar echo data, pre-processing the meteorological radar echo data to obtain a radar echo grayscale image, on the basis of an ST-LSTM unit, constructing a recurrent neural network model with a memory decoupling module, converting the meteorological radar echo grayscale image into a time sequence, and dividing the time sequence into a training set and a test set. The training set is substituted into the recurrent neural network with the memory decoupling module in step S2 for training, and the test set is used for verification in the training process, and the training is stopped when the training frequency reaches a preset frequency. The trained network is used for thunderstorm prediction on the test set, and a short-term near-time thunderstorm prediction image time sequence is obtained. The application is trained to focus on different aspects of space-time changes of thunderstorms in different memory states, does not learn redundant features, and improves the accuracy of thunderstorm prediction to a certain extent.

The application provides a lightning early warning method and system based on wavelet decomposition and an early warning terminal, and relates to the technical field of lightning early warning. The method comprises the following steps: obtaining original atmospheric data in a preset time period; selecting a wavelet information to perform wavelet decomposition on a noisy signal; configuring a preset threshold value for each layer coefficient obtained through decomposition; performing inverse wavelet transformation on the coefficient after noise reduction processing to obtain reconstructed atmospheric data after noise reduction; obtaining a low-frequency component representing an approximate trend of atmospheric electric field change and a high-frequency component representing a detailed part of the atmospheric electric field through 5-layer decomposition of a sym4 wavelet function; zero-filling the d5 coefficient to reconstruct the d5 coefficient component under the condition that the sampled atmospheric data meets the Nyquist criterion; and using the d5 coefficient component to detect the sudden change of the atmospheric electric field and to early warn the lightning, so that the lightning occurring in the thunderstorm passage stage can be early warned, the time point of the thunderstorm early warning can be determined, the lightning early warning requirement can be met, and the early warning precision is improved.

The present application provides a thunderstorm activity observation and management system and method applicable to a power transmission channel. The method comprises: acquiring the number of thunderstorm days in each analysis period; respectively adjusting combined with a preset discrepancy threshold so as to obtain a thunderstorm intensity value of each analysis period; determining a voltage intensity coefficient for each analysis period according to line feature data of a power transmission channel acquired in each thunderstorm activity; and determining a thunderstorm activity assessment value according to the above two result values, wherein the thunderstorm activity assessment value is used for determining a defense level, thereby, defense resources can be scientifically allocated according to the defense level.

An airborne weather radar-based rapid thunderstorm identification method, relating to the technical field of airborne weather radars. The method specifically comprises: performing ground clutter suppression processing on acquired three-dimensional echo data of an airborne weather radar; performing reflectivity factor calculation for data at each beam position during a volumetric scanning process, so as to obtain three-dimensional echo reflectivity factor data of the airborne weather radar after the ground clutter suppression; processing the three-dimensional echo reflectivity factor data after the ground clutter suppression, so as to obtain two-dimensional planar reflectivity factor data; extracting potential thunderstorm areas from the two-dimensional planar reflectivity factor data; mapping data of the potential thunderstorm areas back to the three-dimensional reflectivity factor data, so as to derive thunderstorm cell features; and, on the basis of a thunderstorm identification feature threshold, performing determination on the potential thunderstorm areas and the derived thunderstorm cell features. The method can improve the speed and accuracy of thunderstorm identification, and improve pilots' awareness of hazardous weather areas along flight routes.

This invention belongs to the field of lightning detection technology, specifically relating to a lightning location method and system based on hybrid baselines. It aims to solve the problem that existing technologies cannot integrate short-baseline and long-baseline lightning detection data, thus failing to achieve hybrid baseline lightning location. The invention includes: real-time acquisition, labeling, and compression of lightning electromagnetic pulse waveform data by long- and short-baseline electromagnetic pulse detection devices; real-time transmission of the compressed lightning electromagnetic pulse waveform data to a data processing center station; decompression and filtering of the lightning electromagnetic pulse waveforms at the data processing center station; and calculation of the accurate arrival time of the signal at each detection device at the data processing center station; and calculation of the lightning event location using a time-difference positioning algorithm at the data processing center station. This invention can significantly improve the detection range, efficiency, and positioning accuracy of existing lightning electromagnetic pulse detection systems, and is suitable for low-cost, large-area lightning event detection, especially for accurate real-time detection of thunderstorm activity in offshore areas.

The application discloses a weather radar dynamic grid monitoring meteorological avoidance area boundary extraction method, which is characterized in that: three-dimensional space is gridded, weather radar is used to dynamically monitor thunderstorm reflectivity data of each grid, aircraft flight risk of any grid is calculated, whether each grid is in a safe flight state is determined according to a flight safety threshold of a pilot of different types of aircraft, and all grids are combined into a three-dimensional meteorological avoidance area according to their flight-prohibited states, so that different types of aircraft are guided to fly safely.

This invention belongs to the technical field of high-energy electron deposition discrimination methods, and particularly relates to a discrimination method for high-energy electron deposition induced by low-latitude lightning whistle. The method involves low-latitude lightning whistle sound wave observation and acquisition, obtaining ELF / VLF band electromagnetic wave data recorded by a space observation platform in a low-latitude region, identifying lightning whistle sound wave signals related to continuous lightning activity in thunderstorm areas, and extracting the spectral range, temporal evolution characteristics, and persistence characteristics of the whistle sound waves. The method also includes ray tracing screening of the low-latitude ionospheric penetration frequency band, performing ray tracing calculations on the candidate frequency range of lightning whistles based on low-latitude ionospheric and geomagnetic field environmental parameters to determine the penetrable frequencies that can leak from the Earth-ionospheric waveguide and penetrate the low-latitude ionospheric region, and merging these to obtain the penetrable frequency band interval.

The disclosure provides an electric power transmission safety monitoring system for a thunderstorm area. The system includes a transmission platform, an area characteristic analysis unit and a transmission influence analysis unit, where the transmission platform may receive an electric power transmission request sent by a current power supply end, determine a power demand end and a transmission area where the power demand end is located. After receiving the area characteristic analysis signal sent by the transmission platform, the area characteristic analysis unit carries out thunderstorm characteristic analysis, and sends the analysis result to the transmission influence analysis unit and the transmission platform. The transmission platform further formulates an electric power transmission strategy.

The application discloses a thunderstorm identification method based on dual-polarization radar, and relates to the technical field of thunderstorm identification.The method comprises the following steps: obtaining a thunderstorm reflection threshold based on the reflectivity of a first number of thunderstorms; obtaining a thunderstorm coordinate point based on the thunderstorm reflection threshold, real-time coordinates and real-time reflectivity; obtaining a single thunderstorm point based on the thunderstorm coordinate point; constructing real-time thunderstorm parameters and a fitting midpoint based on the single thunderstorm point; obtaining a pre-warning thunderstorm threshold based on a second number of thunderstorms that need to be pre-warned; judging whether the thunderstorm is a thunderstorm that needs to be pre-warned based on the pre-warning thunderstorm threshold and real-time thunderstorm parameters; and obtaining a thunderstorm horizontal moving direction based on the fitting midpoint.The application is used to solve the problem that the existing thunderstorm identification technology cannot set a suitable reflectivity threshold, simple pre-warning judgment and simple thunderstorm moving direction acquisition, resulting in that the thunderstorm identification result is not fast and accurate enough.

The invention discloses a thunderstorm splitting and merging process detection method and system based on a long-term memory algorithm, and relates to the technical field of meteorological monitoring and early warning, and the method comprises the steps: receiving multi-source meteorological data (Doppler radar reflectivity data and lightning positioning data), carrying out the collaborative preprocessing of the two types of data, and obtaining a multi-source data preprocessing result; obtaining space-time sequence sample effective data adaptive to thunderstorm detection; extracting a long-time sequence labeling sample in the effective data, inputting the long-time sequence labeling sample into a pre-constructed long-time memory network model for training, and outputting the trained long-time memory network model (the model comprises three core stages of feature extraction, feature fusion and feature decoding); and inputting effective data of the space-time sequence sample into the trained long-time memory network model, and outputting a detection result (including starting time, position and dynamic trend) of a thunderstorm splitting and merging process.

The utility model relates to the technical field of photovoltaic power generation monitoring, and discloses a monitoring device for photovoltaic engineering, which comprises a plurality of machine bodies, foot stools fixedly connected to the bottom ends of the machine bodies, displacement mechanisms fixedly connected to the bottom ends of the foot stools, and a light following mechanism fixedly connected to the output end of a third electric telescopic rod. The output ends of the first electric telescopic rods are rotationally connected with connecting shafts, the left connecting shaft is fixedly connected with a wind measuring mechanism, the left side of the right connecting shaft is fixedly connected with a temperature, humidity and pressure measuring mechanism, the right side of the right connecting shaft is fixedly connected with a radiation measuring mechanism, and the radiation measuring mechanism is used for monitoring the radiation value in sunlight. A lightning rod is fixedly connected to the top middle of the right connecting shaft. According to the utility model, the displacement mechanism is used, the whole device can be conveyed to an ideal position, and in addition, the device is provided with the lightning rod, so that the monitoring device is prevented from being damaged by thunderstorm weather.

The invention discloses a clamping type voltage sensor and a lightning protection control method thereof, and relates to the technical field of power system monitoring, the clamping type voltage sensor comprises a first shell, a second shell is hinged to the first shell, the first shell and the second shell are both provided with semicircular grooves used for clamping transmission wires, and when the first shell and the second shell are closed, the first shell and the second shell are closed. The first shell and the second shell are provided with two semicircular grooves, the two semicircular grooves form a circular groove, the first shell and the second shell are assembled and connected with a buckle structure used for enabling the first shell and the second shell to be mutually buckled, a voltage measuring assembly and a power taking assembly are installed in the first shell and the second shell, and the power taking assembly supplies power to the voltage measuring assembly; according to the invention, lightning stroke of thunderstorm weather to the sensor can be effectively prevented, the clamping type voltage sensor cannot slide on the transmission wire through clamping, and the stability of a measurement system is improved.

The invention discloses a grey prediction and comparative learning-based dynamic course change path planning method and system, and belongs to the technical field of air transportation. The method comprises the following steps: identifying a thunderstorm area by using meteorological radar data, and screening an effective area through DBSCAN clustering; a Graham algorithm is adopted to construct a static flight limited area; dynamically predicting the flight limited area at the future moment based on the grey prediction model and expanding the safety margin; multi-modal features are constructed by fusing aircraft states, environmental factors and prediction results, and high-quality feature representation is obtained through comparative learning; and finally, generating a fly-around path by using a PPO reinforcement learning algorithm. The system comprises corresponding function modules. According to the method, the problem of flight revision caused by dynamic change of a flight limited area in severe weather is effectively solved, the accuracy, safety and efficiency of path planning are improved, and the flight delay rate is remarkably reduced.

This application discloses a method and related device for predicting lightning occurrence areas and intensity changes, relating to the field of meteorological monitoring and early warning. The method includes: identifying and tracking radar volume scan data of a target area to obtain thunderstorms; calculating a preset reflectivity peak height and its changing trend within the thunderstorm range; when the preset reflectivity peak height is rising and the electrification height exceeds a preset height, identifying horizontal grid points within the thunderstorm range as potential lightning activity areas; calculating the spatial distribution of potential lightning activity areas in subsequent time intervals; calculating the distribution of the vertical reflectivity core height at each horizontal grid point in subsequent time intervals; determining the lightning activity intensity level at each grid point; establishing a correspondence model between the vertical reflectivity core height and lightning activity intensity; and determining the stage of lightning activity within the potential lightning activity area based on the currently identified core height and its changing trend. This method enables high-precision and high-timeliness near-term lightning warnings.

The utility model discloses a kind of propaganda columns for electrical equipment sales, including propaganda column body, the side of propaganda column body is equipped with rotating shaft by bearing, the rotating shaft is provided with the sheltering mechanism that can be sunshade or rain shelter for visitors, the sheltering mechanism includes mounting rod, sunshade, clamping groove, snap pin, fixed ring, spring, mounting block, pull block, the rotating shaft is fixedly connected with mounting rod, the upper end of the mounting rod is fixedly connected with sunshade, the rotating shaft is equipped with clamping groove, the snap pin is slidably connected in the clamping groove, the utility model is through design mounting rod, sunshade, snap pin, spring and the like component, can be through sunshade to the sun to be shaded, to ensure that the temperature under propaganda column is slightly lower compared with outdoors, and thunderstorm can guarantee that demander and leaflet etc. are not wet, and movable snap pin can remove indirectness to remove the limiting of sunshade, rotate sunshade to make it 90 ° folding, can reduce the volume of this propaganda column, it is convenient to store.

The utility model discloses a kind of lightning protection mechanism's pole circuit breaker belong to pole circuit breaker field, including circuit breaker main body, insulating pillar is installed on circuit breaker main body, and upper terminal is equipped on insulating pillar, lower terminal is equipped on insulating pillar. Through the lightning protection mechanism set, in the process of pole circuit breaker use, can effectively carry out lightning protection, so that pole circuit breaker can also effectively use in thunderstorm weather, avoid high-voltage lightning to cause the damage of pole circuit breaker, adopt shield, while playing lightning protection shield, the main part of pole circuit breaker can be played certain rainproof effect, by the first connecting sleeve, second connecting sleeve, fixed rod and movable rod set, can be connected to the upper end of stand, the effect of auxiliary support to shield is played, increase the stability of use, the inside of movable rod is set to clamping component, wire can be assisted clamping, improve the stability of wire.

The invention provides an offshore floating photovoltaic power station lightning protection system and a construction method thereof, and relates to the technical field of photovoltaic power generation lightning protection, and the system comprises a lightning receiving protection device which is installed at the top of a photovoltaic power station and is used for guiding lightning current; the leading-down system is used for leading the lightning current into the grounding device; the grounding device comprises a ground screen and a vertical grounding electrode which are arranged around the floating body channel, and the vertical grounding electrode is made of an anti-corrosion material and extends downwards from the ground screen to be driven into the seabed; the electromagnetic shielding device is installed in the photovoltaic power station and used for reducing transmission of lightning stroke charges; and the real-time monitoring device comprises thunderstorm early warning equipment, grounding resistance testing equipment and induced voltage monitoring equipment, and is used for dynamically monitoring lightning stroke events and triggering maintenance instructions, so that the technical problems of poor durability and low reliability of the lightning protection system of the ocean floating photovoltaic power station are solved, and the effects of improving the operation safety, stability and maintainability of the lightning protection system are achieved.

The invention relates to the technical field of meteorological disaster early warning, and discloses a thunderstorm early warning model training method and device and a thunderstorm early warning method and device, and the thunderstorm early warning model training method comprises the steps: obtaining a historical thunderstorm data set of a to-be-predicted region; the thunderstorm record data of each thunderstorm are analyzed to obtain thunderstorm features of each thunderstorm, the thunderstorm features comprise space thunderstorm features at multiple moments and time sequence thunderstorm features at multiple feature positions, and the space thunderstorm features are obtained according to the distance between each monitoring station and a thunderstorm centroid coordinate; the time sequence thunderstorm features are constructed according to the monitoring data sets of the feature positions at different moments and the monitoring data sets of all the monitoring stations, and the time sequence thunderstorm features are constructed according to the monitoring data sets at different moments; and establishing a training set according to the thunderstorm features of the thunderstorms and the thunderstorm early warning parameters, and training a preset model by using the training set to obtain a thunderstorm prediction model. According to the invention, through double analysis of the thunderstorm data in time sequence and space, the accuracy of thunderstorm prediction is increased, and then the thunderstorm early warning efficiency is improved.

The utility model belongs to the technical field of zero-carbon containers, particularly relates to a zero-carbon container type building device, and aims to solve the problems that in the prior art, in thunderstorm weather, a solar panel is exposed outside for a long time, the service life of the solar panel is easily shortened, and water resources cannot be collected and used. The container comprises a container body and a recycling box arranged on the top of the container body. The protection mechanism is arranged in the recycling box; the rainwater collecting mechanisms are arranged on the two sides of the recycling box and used for collecting rainwater; the device comprises a rainwater collecting mechanism, a recycling mechanism and a protection mechanism, the recycling mechanism is arranged on one side of the rainwater collecting mechanism, the protection mechanism comprises a hydraulic cylinder arranged in the center of the bottom side in a recycling box, and the output end of the hydraulic cylinder is fixedly connected with a solar panel. And resource waste is prevented through collection and use.