167 results about "Thunderstorm" patented technology

The application discloses a thunderstorm monitoring device and path planning method of unmanned aerial vehicle based on electric field and magnetic field detection. The device comprises a quadrature magnetic antenna array, a vertical differential electric field sensor, a signal processing module, a positioning algorithm module, a distance estimation module, an autonomous decision module, a GNSS module and an inertial measurement module. The quadrature magnetic antenna array is composed of two orthogonally arranged magnetic antennas, which are at a 45° angle with the flight direction of the unmanned aerial vehicle, and is used for receiving the horizontal magnetic field signal of lightning radiation. The vertical differential electric field sensor comprises two groups of upper and lower electrode plates and measures the differential value of the vertical electric field intensity. Through the cooperative measurement of the magnetic antenna array and the electric field sensor, the lightning azimuth (θ) is calculated by combining the magnetic field intensity ratio, and the quadrant is corrected by using the electric field polarity. Based on the magnetic field amplitude attenuation model and the waveform rise time correction factor, a distance estimation model is constructed. The autonomous decision module generates an approach path or an obstacle avoidance path according to the lightning azimuth and distance, and drives the flight control system to adjust the heading.

The invention provides a weather forecasting method and system for thunderstorm and gale. The method comprises the following steps: acquiring multi-source weather observation data in a past preset time period; inputting the multi-source meteorological observation data into a pre-trained progressive wind speed forecasting model to extract and fuse multi-scale spatial-temporal characteristics in the multi-source meteorological observation data, and sequentially generating average wind speed forecasting results covering a plurality of future time periods; inputting the average wind speed forecast result into a pre-trained gust mapping model, and performing nonlinear mapping to obtain gust wind speed forecast results corresponding to a plurality of future time periods; and based on the gust speed forecast result, whether thunderstorm and gale risks exist in the future time period is judged. In the mode, the multi-source meteorological observation data is acquired and prediction is performed in combination with the progressive wind speed prediction model and the gust mapping model, so that the accuracy of wind speed and gust prediction can be improved, the thunderstorm and gale risk can be identified in advance, and the disaster early warning and disaster prevention and reduction capabilities are further improved.

The invention discloses a TITAN product thunderstorm forecast field correction method and system based on multi-scale ensemble learning. The method comprises the following steps: S1, forming a unified multi-source meteorological data set and a TITAN product set; s2, constructing a meteorological target dichotomy label; s3, preprocessing the multi-source meteorological data set, and matching meteorological target dichotomy tags to generate a sample data set; s4, calculating a multi-scale feature set reflecting macro and micro features of the thunderstorm system and the non-thunderstorm system based on the sample data set; s5, generating a comprehensive multi-scale feature expression; s6, forming a plurality of sub-models with differentiated scale sensing capability; s7, training to form a final thunderstorm system identification model; s8, constructing a TITAN product forecast field reconstruction module; and S9, generating a forecast result reflecting the spatial distribution, the development trend and the potential influence area of the thunderstorm system in real time. According to the method, the time-space evolution feature capturing precision of the thunderstorm system and the prediction stability and reliability are improved.

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 is suitable for the technical field of thunderstorm prediction, and provides a strong thunderstorm potential risk and intensity prediction method based on an artificial neural network, and the method comprises the steps: obtaining the ground-to-ground lightning data of a target region, determining the time distribution characteristics of a thunderstorm event, and recognizing the time valley of the thunderstorm event; collecting multi-source meteorological data in a target area, and screening meteorological prediction factor data; performing space-time alignment processing on the ground-to-ground lightning data and the meteorological prediction factor data, and constructing a grid unit day-by-day sample set based on the aligned data; constructing a probability classification model of a strong thunderstorm event based on the grid unit day-by-day sample set; building a regression neural network model based on the grid units which are judged to be strong thunderstorm high risks by a probability classification model; and a collaborative prediction result is output, a predicted value of a geometric mean value of the ground-to-ground flash frequency and the lightning current amplitude is synchronously output, and a risk assessment basis of classification discrimination and numerical prediction capability is provided for extreme weather early warning and power grid disaster prevention and reduction through double-order modeling.

The invention provides a rainstorm lightning disaster risk early warning method and system based on an analytic hierarchy process, and the method comprises the steps: obtaining historical thunderstorm frequency data and thunderstorm movement path data of a target region, and analyzing and determining the sensitivity levels of different regions in thunderstorm activities; the lightning stroke risk assessment model is decomposed into a target layer, a criterion layer and a scheme layer by using an analytic hierarchy process; according to historical data including lightning stroke historical data, thunderstorm historical data and disaster damage historical data, performing pairwise comparison on each factor of the criterion layer, and constructing a judgment matrix; acquiring spatial position information and characteristics of different types of disaster-bearing bodies, performing overlay analysis according to spatial positions of the disaster-bearing bodies and the lightning drop point probability distribution map, calculating lightning stroke probability values of areas where the disaster-bearing bodies are located, and generating a disaster-bearing body risk level distribution map.

The invention relates to the technical field of thunder and lightning early warning, and discloses a thunderstorm early warning system and method applied to a new energy station, and the system comprises a data collection module, a data processing module and an early warning decision module which are arranged in a target new energy station. The multi-source thunderstorm early warning information is sent to a data processing module; wherein the multi-source thunderstorm early warning information comprises lightning data, radar data, satellite data, micrometeorological data and atmospheric electric field data, and the data processing module performs feature extraction on the multi-source thunderstorm early warning information to obtain a first thunderstorm early warning feature vector and a second thunderstorm early warning feature vector; and the early warning decision module performs thunderstorm early warning based on the first thunderstorm early warning feature vector and the second thunderstorm early warning feature vector to obtain thunderstorm early warning information of the new energy station. According to the invention, the accuracy of the thunderstorm early warning information of the new energy station is improved.

The invention discloses a lightning multi-parameter space-time analysis method based on clustering tracking and a thunderstorm risk early warning method and system, and relates to the technical field of severe convective weather early warning. The method comprises the following steps: collecting and analyzing lightning data; data preprocessing and space-time collaborative clustering: clustering lightning belonging to the same thunderstorm monomer in the current time window by using optimized clustering parameters; cross-period target tracking and association: constructing a lightning cluster tracking chain of a continuous time window; performing multi-factor dynamic feature extraction, namely capturing real-time dynamic feature parameters on the associated cluster sequence; and constructing a hazard index model and carrying out risk early warning. According to the method, dynamic tracking, strength evaluation and danger grading of thunderstorm activities can be realized by utilizing lightweight and easily acquired lightning positioning data, the method has the characteristics of light calculation load and high real-time performance, and objective and quantitative technical support can be provided for refined short-term and temporary early warning decision making.

The invention relates to an intelligent thunderstorm weather forecasting method and system and a medium, belongs to the field of thunderstorm forecasting, and provides the intelligent thunderstorm weather forecasting method for solving the problem that an existing mode is insufficient in forecasting capacity, and the intelligent thunderstorm weather forecasting method comprises the following steps: collecting and processing initial data, and obtaining a multi-source fusion initial field of a unified format; constructing a short-time forecasting model based on a Swin Transform backbone network, and outputting a thunderstorm forecasting field in the future 6 hours by inputting a multi-source fusion initial field and underlying surface information; and a short-term forecasting model based on a FuXi large model is constructed, and by inputting the multi-source fusion initial field and underlying surface information obtained in the step 1, possible thunderstorm occurrence areas and intensity distribution data in more than six hours in the future are output. By introducing a Swin Transform backbone network and combining an autoregression prediction strategy and a multi-task loss function, the capturing precision and real-time performance of the thunderstorm time-space evolution process are effectively improved.

The invention discloses a hail cloud early recognition method based on a polarization radar. The method comprises the following steps: establishing a sample library; extracting a plurality of decision factors of hail cloud monomer radar data 20 minutes before hail occurs in the sample library, establishing a logistic regression model for the plurality of decision factors in the sample library, selecting an optimal weight algorithm by using a cross validation mechanism, calculating weights of the plurality of decision factors, and determining a hail early warning index threshold; the early warning index is compared with the hail early warning index threshold value, if the early warning index is larger than or equal to the hail early warning index threshold value, it is judged that the hail cloud is an early-stage hail cloud monomer, and if not, it is judged that the hail cloud is a general thunderstorm cloud. Based on polarization radar data and three key judgment factors, the early recognition accuracy of hail clouds is effectively improved, the hail clouds and general thunderstorm clouds can be accurately distinguished, and weather forecast and manual hail suppression operation are assisted.

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 thunderstorm cloud electric field prediction method, device and equipment based on a deep learning algorithm, and relates to the technical field of thunderstorm cloud electric field prediction. The method comprises the following steps: firstly, acquiring a target convection parameter group of a region of interest, and acquiring historical observation data of each convection parameter in the target convection parameter group and historical observation data of an electric field in a thunderstorm cloud; constructing a convection parameter prediction model and a thunderstorm cloud internal electric field inversion model based on historical observation data; integrating the two models to obtain a thunderstorm cloud internal electric field prediction model corresponding to each convection parameter set; and performing thunderstorm cloud internal electric field prediction on the region of interest according to the thunderstorm cloud internal electric field prediction model corresponding to each convection parameter set. According to the application, convection parameter prediction and in-cloud electric field inversion are fused, and prediction of the thunderstorm in-cloud electric field can be realized by using observable convection parameters.

The invention discloses a 750kV power transmission line plateau type integrated lightning protection method and system, and relates to the technical field of power systems, and the method comprises the steps: collecting the data of a plateau region through the Internet of Things to construct a plateau data set, carrying out the lightning correlation feature screening of the plateau data set, obtaining lightning features, and carrying out the preprocessing of the lightning features, and obtaining pure features; the method comprises the following steps: collecting data of a plateau region through the Internet of Things to construct a plateau data set, performing lightning strike associated feature screening on the plateau data set to obtain lightning features, and preprocessing the lightning features to obtain pure features; establishing a comprehensive evaluation model, and inputting the pure features into the comprehensive evaluation model to obtain a feature comprehensive score; constructing and training a thunderstorm prediction model based on LSTM, and inputting the feature comprehensive score into the thunderstorm model to obtain a thunder and lightning activity prediction value; and dividing a risk interval for the lightning activity prediction value, and making a corresponding lightning protection measure according to the risk interval.

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 invention discloses an agricultural unmanned aerial pesticide application method based on a real-time dynamic positioning technology, and belongs to the technical field of unmanned aerial vehicles. The method comprises the following steps: 1, AI atmosphere prediction; 2, intelligent mode switching; and multi-source data fusion is fused into the two methods, and the multi-source data is preprocessed for time alignment, spatial interpolation and 3 sigma principle abnormal value elimination. Wherein in the RTK + IMU tight coupling mode, IMU mechanical arrangement data, RTK observation values, 15-dimensional state vector modeling poses and sensor errors are fused through Kalman filtering; and in a LiDAR SLAM mode, the point cloud is subjected to ground removal through RANSAC, and after feature extraction, a digital twin map is matched, so that GNSS-free environment centimeter-level positioning is realized. After the technologies are introduced into the unmanned aerial vehicle, the positioning error of the unmanned aerial vehicle in thunderstorm days can be greatly reduced, the positioning availability of the unmanned aerial vehicle in complex terrains covered by signals reaches 98% or above, and the working efficiency in complex regions can be remarkably improved.

The invention provides an active lightning triggering device and method for a fan blade. The device comprises two copper rods with tip heads, a protection circuit, a boost lightning circuit and a metal shielding net. The two copper rods are arranged in parallel at the blade tip of the fan blade, one copper rod is directly grounded, and the other copper rod is connected with the boost lightning circuit through the protection circuit; the boost lightning guide circuit comprises a charging capacitor, an inductor, a high-frequency switch and a diode and is used for sensing positive high voltage and dynamically adjusting resonance characteristics so as to enhance voltage difference between thundercloud and a copper rod and trigger uplink and downlink leader to form a discharge channel; the metal shielding net wraps the blade tip and is connected with the grounding device to ensure that lightning current is preferentially discharged through the copper rod, and the blade tip structure is prevented from being damaged. According to the method, environmental meteorology and blade attitudes are monitored in real time, a boosted circuit is started in combination with threshold judgment, and control parameters are optimized by using a remote monitoring platform, so that intelligent lightning protection management is realized. According to the device, by actively exciting corona discharge and accurately guiding a lightning path, the lightning protection efficiency of the blade is remarkably improved, the damage risk of a non-lightning-receiving area is reduced, meanwhile, a maintenance decision is supported through datamation operation and maintenance, and the device is suitable for wind generating set protection in a thunderstorm area.

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 application provides a lightning protection monitoring method and system, the method comprises the following steps: obtaining past atmospheric weighted temperature data set, past thundercloud observation data set and past radar echo data set to train and construct an initial thundercloud prediction neural network, and then obtaining a final thundercloud prediction neural network; obtaining a plurality of predicted thundercloud maps based on the final thundercloud prediction neural network, converting the predicted thundercloud maps into thundercloud grid maps, and obtaining a plurality of thunderstorm grid areas in the thundercloud grid maps; obtaining a predicted thunderstorm moving path based on the thunderstorm grid areas in the thundercloud grid maps in adjacent time frames to perform lightning protection. The final thundercloud prediction neural network is used for preliminary thunderstorm prediction, the predicted thunderstorm moving path is obtained through the predicted thundercloud map, the accurate prediction result of the position and time of the thunderstorm is obtained, and then the lightning protection accuracy and the lightning protection effect are improved through the combination with passive lightning protection measures.

This invention discloses a fully automated, three-dimensional active defense system and method, belonging to the field of weather modification technology. The system monitors thunderstorm weather systems via airborne remote sensing satellites, triggering an early warning signal when the system moves to the outermost layer of the defense area. It combines this with a network-based ground-based radar monitoring system to continuously track and monitor thunderstorm cloud weather systems in real time, simultaneously sending various parameters to a lightning monitoring and early warning module. This provides criteria for detecting strong lightning electromagnetic fields, inverts the spatiotemporal parameters of lightning occurrence, and provides early warning and forecasting of the development trend of thunderstorm weather systems. A central control system then determines lightning dissipation and initiates operational commands. Finally, a mobile artificial lightning induction system is used to artificially intervene in the charge structure of thunderstorm clouds. This invention can mitigate the direct and indirect electromagnetic damage effects of strong lightning caused by the spatiotemporal unpredictability of severe convective weather and the instantaneous nature of strong thunderstorm discharge processes, protecting the safety of key locations and important sensitive targets.

The invention relates to a lightning protection transformation point identification method and device, equipment, a storage medium and a program product. The method comprises the following steps: receiving an identification request of a lightning protection transformation point; acquiring position data and height difference data of each candidate point in the cable line in the target area, wherein the height difference data is used for representing the height difference between the corresponding candidate point and the highest point of the environment where the candidate point is located; inputting the position data and the height difference data of each candidate point into a thunderstorm risk model to obtain thunderbolt risk prediction data of each candidate point; the thunderstorm risk model is used for performing product weighting on the features corresponding to the position data and the features corresponding to the height difference data, and determining thunderbolt risk prediction data according to the combined features after product weighting; and according to the thunderbolt risk prediction data of each candidate point, determining a lightning protection transformation point of the target area from each candidate point. By adopting the method, the potential lightning protection transformation point can be identified in advance before the thunderstorm fault occurs.

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.

This invention discloses a transformer lightning protection insulation system, comprising: a fast voltage monitoring circuit; a multi-level surge arrester assembly; a microsecond-level response processor for executing a lightning waveform feature detection algorithm, a gradient protection barrier design algorithm, and a microsecond-level response control algorithm; and a protection control execution unit. The lightning waveform feature detection algorithm executed by the microsecond-level response processor is used to accurately model the time-domain characteristics of the lightning current; the gradient protection barrier design algorithm executed by the microsecond-level response processor is used to calculate the protection voltage distribution and current shunting allocation of the multi-level surge arresters; and the microsecond-level response control algorithm executed by the microsecond-level response processor is used to optimize the overall system response time and implement a hierarchical protection strategy. This invention achieves the technical objectives of strong protection capability, microsecond-level fast response, and uniform energy distribution for transformers during thunderstorms.

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 invention provides a method and a system for planning a flight changing path based on a multi-scale algorithm, and belongs to the technical field of flight path planning, a multi-scale map with a customizable maximum depth is constructed, and the safety margin of the map is increased through edge detection and an expansion algorithm. The multi-scale map can efficiently realize three functions of position-node mapping, adjacent node search and visibility judgment. Global path planning is completed based on an A * algorithm, and the current predicted thunderstorm area is avoided. And according to the navigation real-time weather data, determining a navigation change starting point and generating a plurality of path schemes. And arranging voyage and safety margin data, calculating index weights by adopting a fuzzy comprehensive evaluation method and combining an analytic hierarchy process, and performing comprehensive evaluation on the scheme. And preferentially selecting the highest score scheme. According to the method, the search efficiency is improved through a multi-scale map, dynamic planning and comprehensive evaluation are combined, the path safety and the navigation easiness are ensured, efficient and reliable navigation changing path planning is achieved, and dynamic meteorological conditions such as thunderstorm are effectively coped with.

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.