620 results about "Torrential rain" patented technology

Disclosed in the present invention is an intelligent generation method for urban waterlogging point-surface combined continuous monitoring data based on multi-modal data. The method comprises the following steps: 1) collecting multi-modal data from a plurality of sources to form a multi-modal data set; 2) intelligently extracting rainstorm flood information on the basis of a deep learning algorithm; 3) matching the rainstorm flood information on a spatio-temporal scale; 4) generating spatio-temporally continuous rainstorm flood data by means of a generative adversarial network algorithm; and 5) using a comprehensive evaluation method for deduplication processing to generate an urban waterlogging point-surface combined continuous monitoring database based on multi-modal data. In the present invention, on the basis of conventional rainstorm flood monitoring data, multi-modal data collected from the plurality of sources is supplemented, rainstorm flood information contained in the multi-modal data is intelligently extracted on the basis of the deep learning algorithm, the urban waterlogging point-surface combined continuous monitoring database based on the multi-modal data is constructed, and the defects that conventional monitoring data is scarce and has fixed sources are overcome.

The invention discloses a non-point source pollution emergency bypass decision control method, particularly relates to the technical field of automation and process control, and aims to solve the problems that existing scheduling cannot align multi-source measurement and evaluate credibility in minutes under a unified time base, and is lack of risk measurement integrated with a tide level phase and a moisture regain path and same-layer gating. The bypass rhythm is easy to misjudge, and the oscillation is difficult to audit. A minute-level risk density is constructed through multi-source alignment and credibility grading under a unified time reference, injected tide level phases and moisture regaining paths, bypass rhythms are constrained by same-layer gating and tide windows, and beam search, boundary arbitration and fixed window closed-loop correction are matched, so that the risk density is improved. Therefore, under the superposition of rainstorm and tide, decision making and execution are stably limited in a non-backflow red line, bearing and equipment limit, misjudgment and oscillation are remarkably reduced, the problems of unavailability and difficult auditing caused by multi-source asynchronization and moisture regain coupling are solved, intelligent decision support is provided for non-point source pollution treatment, and it is ensured that the treatment technology is accurate and efficient.

The invention discloses a rainstorm torrential flood physical constraint prediction method and system based on a graph neural differential equation, and belongs to the technical field of rainstorm torrential flood prediction. Carrying out space-time attention fusion based on a graph; carrying out modeling and dynamic deduction based on a graph neural differential equation of physical constraints; predicting and outputting a multi-task flood hydrograph; and carrying out joint loss function design and end-to-end training. The system comprises a multi-modal hydrological feature obtaining and coding module used for multi-source heterogeneous data feature extraction, a graph-based space-time attention fusion module used for deep fusion of multi-source heterogeneous features, and a physical constraint-based graph neural differential equation dynamic core module used for continuous dynamic process modeling. And the prediction output module is used for outputting the spatial distributed flood hydrograph. According to the method, the problems of low reliability, poor timeliness and poor extrapolation capability during rainstorm torrential flood prediction in the prior art are solved.

The invention provides a vehicle control method and device for an intelligent driving scene, electronic equipment and a medium. The method comprises the steps of obtaining multi-mode information collected by multiple sensors installed on a vehicle, vehicle system fault diagnosis information and a current driving environment; on the basis of the multi-modal information, the confidence degree of each piece of modal information in the current driving environment is calculated; according to the confidence degree of each piece of modal information, performing feature fusion on the multi-modal information to obtain an environment state vector; and generating a vehicle control strategy based on the environment state vector, the system fault diagnosis information and the current driving environment. According to the method, by introducing a confidence coefficient evaluation mechanism, unreliable modes affected by environment interference or faults can be automatically inhibited, and the perception accuracy and stability under complex scenes (such as night rainstorm and strong backlight) are remarkably improved.

The invention provides an automatic driving test scene generation method based on real traffic data, and solves the problems of low accident data utilization rate, SIL / HIL test splitting and insufficient boundary coverage in the prior art. Comprising the following steps: acquiring multi-source heterogeneous traffic accident data; cleaning data by adopting a joint interpolation-anomaly detection mechanism; vehicle dynamic sudden change characteristics within 0.5 second before braking are extracted through LSTM and DTW algorithms; constructing a three-dimensional scene pipeline driven by a physical engine, and dynamically associating the pavement slippery coefficient with the rainfall intensity; analyzing the accident text into simulation parameters by using a semantic-physical parameter converter; performing SIL-HIL cooperative verification: performing extreme illumination perception test and narrow road planning verification in an SIL environment, and realizing 1ms step length fault injection test in an HIL environment; positioning failure parameters based on Bayesian optimization; a GAN is adopted to generate a long-tail scene, and a test boundary is expanded by coupling extreme conditions such as rainstorm / low visibility; and outputting a standard scene library containing the collision probability thermodynamic diagram. The safety verification efficiency under the extreme working condition is remarkably improved.

The invention discloses a flood type landslide disaster monitoring and early warning method, and belongs to the technical field of geological disaster prediction. The method comprises the following steps: step 1, acquiring a flood type landslide disaster case in which a rainstorm event and a landslide event coincide in time and space, collecting landslide factor data, environment data, monitoring data and historical landslide data, and constructing a historical database; step 2, constructing and training a Bayesian network model based on a historical database; step 3, acquiring landslide monitoring data and preprocessing the landslide monitoring data; step 4, training an LSTM time sequence prediction model; 5, inputting the real-time monitoring data into the LSTM model, and predicting to obtain future landslide monitoring data; and step 6, inputting future monitoring data into the Bayesian network to obtain a slope instability probability based on a future trend. According to the invention, through organic fusion of the Bayesian network and the LSTM, dynamic prediction and real-time response of the landslide instability probability are realized, and timeliness, accuracy and robustness of early warning are significantly improved.

The invention relates to the technical field of disaster early warning, and discloses a rainstorm early warning method based on multi-source forecast product dynamic fusion, which comprises the steps of collecting multi-mode rainfall forecast data, constructing a rainfall cumulative distribution function, performing dynamic updating based on Kalman filtering, and obtaining frequency correction rainfall forecast products of each mode; historical forecast and live monitoring data samples are obtained, comprehensive weight coefficients corresponding to all levels of rainfall are calculated, and a fusion initial rainfall distribution field is obtained; performing matching reconstruction on the corrected rainfall forecast field to generate a reconstructed rainfall distribution field; collecting latest real-time monitoring data, and carrying out space-time dynamic correction on the reconstructed rainfall distribution field to generate a real-time rainstorm potential field; and comparing the real-time rainstorm potential field with the multistage rainstorm early warning threshold, generating rainstorm early warning information, and issuing the rainstorm early warning information. According to the invention, the timeliness and accuracy of rainstorm early warning response are improved, the public personal safety is guaranteed, and the city emergency disposal capability is improved.

The invention relates to the technical field, and discloses a power distribution network multi-time scale fault scene deduction method comprising the following steps: obtaining meteorological and power grid operation data, establishing a time sequence fault tree model, and analyzing the trigger probability of multi-line disconnection and rainstorm short circuit faults in a target time period; extracting cross-level fault propagation features, analyzing a dynamic coupling relationship among multi-line disconnection, transformer substation flooding and cascading trip, forming a fault feature mode set, and identifying a single fault and a cascading fault in combination with time sequence analysis; historical fault data are analyzed, time sequence features and topological features of single and cascading faults are extracted, if a single fault propagation path is in a single level, a support vector machine is used for being combined with the features to judge fault types, and a classification result is output; and performing anomaly detection and confidence evaluation on a fault classification result, outputting a fault evolution path and risk evaluation, and analyzing the contribution degree of cascading trip to a large-area power failure risk in combination with historical blackout data to obtain power failure risk probability distribution.

The invention relates to the technical field of mine environment risks, and discloses a mine environment risk multi-modal analysis and early warning decision-making method, which comprises the following steps of: performing time and space reference alignment on multi-source observation data, constructing a unified spatio-temporal data set, and establishing a troposphere disturbance model to realize observation disturbance coupling. And constructing a physical evolution operator and a rainfall-driven external input operator based on a seepage mechanical relationship, and generating a state evolution mechanism. Semantic direction parameters are generated by mapping semantic query information to a target attention dimension, and a lagging propagation operator is constructed by setting a rainstorm ending moment as a starting point. And determining a non-regular spectrum transition index and a peak lag moment through the non-regularity measurement and the dynamic change rate of each lag propagation operator, and setting a reference threshold value based on the spectrum transition index of a historical non-risk time period. Finally, when the early warning intensity meets the condition, early warning is output, and a query result is generated in combination with the lagging propagation operator and the semantic direction parameters.

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 discloses a storm flood disaster emergency decision system and method based on domain knowledge ontology, and relates to the technical field of sudden disaster emergency management and artificial intelligence crossing. The system comprises a domain knowledge ontology module, a multi-agent decision module, a data interaction module and a visual output module. The domain knowledge ontology module is connected with the multi-agent decision module through a knowledge interface; each agent in the multi-agent decision module transmits information through a communication protocol; the data interaction module is respectively connected with an external data source and a multi-department service system, and the data interaction module is connected with the multi-agent decision module through a data transmission channel; the method is suitable for the whole process of monitoring analysis, early warning and forecasting, response processing and evaluation optimization of rainstorm and flood disasters, and through structured knowledge reasoning and multi-agent cooperation, a predictive intelligent decision scheme is provided for disaster prevention and reduction and emergency management departments.

The invention relates to the technical field of hydrometeorological monitoring, in particular to a rainstorm flood water line early warning and monitoring method. According to the invention, the consistency and timeliness of real-time water level data from different data sources are ensured through a multi-source hydrometeorological data acquisition and time synchronization technology. Through accurate time synchronization processing, deviation caused by data time lag is avoided, accurate butt joint and fusion of various water level data are guaranteed, and subsequent flood risk assessment and early warning are more reliable. Particularly, under sudden weather conditions such as rainstorm and the like, quick response can be realized, and accurate risk assessment can be provided; according to the method, the noise in the water level data of each data source is effectively removed by combining a combined preprocessing algorithm of moving average filtering and Kalman filtering, and weights of different data sources are weighted according to prediction errors, so that comprehensive accurate water level data is obtained. According to the fusion method, the smoothness and accuracy of the data are improved, and high-quality input data are provided for subsequent risk assessment.

The invention discloses a weather big data-based rainstorm and flood composite disaster early warning method, which comprises the following steps of: extracting periodic characteristics and abnormal fluctuation information of rainstorm frequency by acquiring long-term weather data and performing hierarchical processing, and analyzing and determining the change trend of the rainstorm frequency in combination with a time sequence. And constructing an incidence matrix based on the rainstorm frequency change trend and the topographic hydrological characteristic data, determining an initial parameter range of flood probability assessment, introducing dynamic correlation analysis, fusing meteorological data and environmental conditions, and obtaining a dynamic adjustment coefficient of probability assessment. Through mapping the adjustment coefficient and the rainstorm frequency change trend, a quantitative expression of a rainstorm and flood coupling structure is constructed, and an early warning model is obtained based on the quantitative expression. And finally generating multi-scene early warning signal distribution by performing adaptive adjustment on the model response and verifying the accuracy of the model response. According to the invention, accurate early warning of flood disasters in a complex environment is realized, and the timeliness and reliability of early warning are significantly improved.

The invention provides a subway tunnel water disaster risk evaluation method under extreme rainfall based on multi-source data, and belongs to the technical field of urban rail transit disaster prevention. The method comprises the following steps: establishing a grading evaluation system on the basis of field rainstorm monitoring data, hydrogeological parameters, field geological exploration, rock-soil physical and mechanical parameters, monitoring data of a tunnel lining structure and material strength parameters of a tunnel site area in combination with a plurality of developed tests, dividing corresponding grades and determining an influence index-; and S4, respectively carrying out weight calculation on the influence index-by adopting an analytic hierarchy process and an entropy weight method, then calculating a dynamic weight of the influence index-, correcting an abnormal dynamic weight, generating a comprehensive score, dividing risk grades, matching graded prevention and control measures, and optimizing and adjusting the prevention and control measures in real time through a parameter risk grade dynamic judgment and construction method matching mechanism. According to the method, minute-level accurate interpretation of a disaster evolution mechanism and dynamic optimization adaptation of a prevention and control scheme are realized.

The invention discloses a multi-scale full-period rainfall flood toughness evaluation method, equipment, a medium and a product, and relates to the field of rainfall flood toughness evaluation. The method comprises the following steps: firstly, according to a rainstorm flood action period and three spatial scales of drainage basin-city-drainage partition, constructing a multi-scale full-period rainfall flood toughness factor library; collecting and calculating each element data in the element layer in the whole cycle process of coping with the rainstorm flood effect under three spatial scales of drainage basin-city-drainage partition; screening out a final multi-scale full-period rainfall flood toughness element library based on the correlation between the elements; calculating the weight of each element in the final multi-scale full-period rainfall flood toughness element library by using an entropy weight method; on the basis of weight calculation, a VIKOR multi-criterion decision-making method is used to calculate rainfall flood toughness under each spatial scale, and a geographic detector is used to analyze elements with most significant influence from two aspects of the spatial scale and a rainstorm flood action stage, so that multi-scale full-period scientific, efficient and accurate evaluation of rainfall flood toughness is realized.

The invention discloses an urban rainstorm waterlogging rapid simulation method based on a deep convolutional neural network, relates to the field of urban waterlogging management, and solves the problem that the judgment of the urban waterlogging prevention capability is inaccurate due to the fact that the current urban waterlogging prevention capability is generally obtained according to the urban water supply and drainage capability and the historical peak rainfall condition. The method comprises the steps that a catchment area corresponding to a target city is set according to a city design drawing, and catchment grids corresponding to the catchment area are divided based on the catchment area; a height-area curve corresponding to the catchment area is constructed according to the catchment grid, and an overflow model corresponding to the catchment area is constructed based on the height-area curve and improved; the result is simulated and recognized through the improved overflow model, then the result is compared with a historical actual value, and the precision condition of the improved overflow model is judged; real-time rainfall data is obtained and imported into the improved overflow model, the simulated submerging conditions of different catchment grids in the target city are obtained, and accurate simulation of urban rainstorm waterlogging is achieved.

The invention discloses a rainstorm ground waterlogging prediction method, device, equipment and product, and relates to the technical field of urban waterlogging prediction.The rainstorm ground waterlogging prediction method comprises the steps that ephemeris and ground observation station data are obtained, the atmospheric moisture delay amount and the high-altitude water vapor conversion factor are calculated respectively, the historical water vapor content is calculated, and the future atmospheric precipitable water amount is predicted; rainfall time distribution data are obtained through rainstorm intensity decomposition; calculating the rainwater overflow amount of the catchment area in combination with the water discharge amount time distribution data of the urban drainage pipe network; and iterative calculation is carried out by using a digital elevation model grid to obtain the waterlogging depth, and waterlogging early warning information is sent out. By constraining the total rainfall amount, the rainfall amount can be effectively constrained by predicting the precipitable water amount based on the satellite ephemeris; the underground pipe network drainage is restrained, and a water flow condition in a pipe channel is simulated by adopting an SWMM model; and finally, the waterlogging accumulated water volume is calculated, constraint verification is carried out on each link of urban waterlogging, and relatively high precision and relatively long timeliness are ensured.

The invention discloses a subway transport capacity matching and passenger flow collaborative management and control method based on multi-agent reinforcement learning, and relates to the technical field of urban rail transit operation management. The method comprises the following components: S1, a subway system digital twin model construction step, S2, a multi-agent reinforcement learning training environment construction step, S3, an extreme scene simulation and agent training step, S4, a virtual-real data interaction and model iteration step and S5, an actual system deployment and decision execution step. According to the method, a high-fidelity digital twin model of a subway system is constructed, extreme scenes including passenger flow retention caused by rainstorm and instantaneous large passenger flow of a large competition field are simulated in the model, and a multi-stage sampling method based on scene clustering is combined to perform training for ten millions of times; and the decision-making capability and the cooperation efficiency of the multiple agents in the face of complex and extreme conditions are obviously improved.

The invention discloses a rainstorm disaster early warning and alarming method, which comprises the following steps: carrying out multi-source data acquisition, and carrying out cleaning, standardization and fusion processing on the acquired multi-source data; based on the basic geographic information data and the historical disaster data, constructing a regional disaster bearing capability evaluation model, and respectively setting differentiated early warning thresholds for regions of different risk levels; based on the preprocessed multi-source data, constructing a rainstorm trend prediction model, and predicting rainfall intensity, cumulative rainfall and rainfall duration in a future preset time period; starting a corresponding multi-channel collaborative alarm mechanism according to the characteristics of the early warning level and the early warning area; in the early warning period, real-time monitoring data are continuously collected, and the rainstorm trend prediction result is updated. According to the method, the problems of inconsistency of multi-source data and low fusion precision are effectively solved, meanwhile, the situation that the prediction advance of secondary disasters such as excessive release of emergency resources and landslide is increased by 1-2 hours is avoided through the temporary high-risk labels, and comprehensive early warning of the secondary disasters is achieved.

The invention provides an autonomous operation obstacle avoidance method and system for a low-altitude aircraft, and the method is executed by the low-altitude aircraft, and comprises the following steps: S1, data acquisition: collecting data through a multi-source heterogeneous sensor group carried by the low-altitude aircraft, airspace control information, real-time meteorological data and collaborative sensing data of other aircrafts sent by the ground sensing base station are received through the air-ground communication interface; preprocessing the collected original data, wherein the preprocessing comprises data format standardization, abnormal value elimination and time synchronization; and S2, data fusion processing: carrying out credibility evaluation on the preprocessed multi-source heterogeneous sensing data and collaborative sensing data. According to the invention, through three-dimensional credibility evaluation and an air-ground data compensation mechanism, the perception precision is greatly optimized, the perception data efficiency is improved in complex environments such as low illumination, strong electromagnetic interference, rainstorm and the like, and obstacle avoidance decision errors caused by data distortion of a single sensor are avoided.

The invention discloses an inland area TRP feature recognition and evaluation method based on data driving and numerical simulation, a medium and a program product, and relates to the technical field of meteorological disaster monitoring and early warning and numerical simulation. And objectively identifying a TRP event which occurs outside the typhoon circulation and is connected with the strong water vapor conveying belt by combining the typhoon peripheral circulation radius, the distance constraint and the whole-layer water vapor conveying flux. Key physical fields such as whole-layer water vapor flux, water vapor income and expenditure, dry invasion and monsoon water vapor surge are diagnosed by utilizing a regional numerical mode and reanalysis data, and a characteristic index set reflecting typhoon, monsoon and dry and cold air coupling influence is formed; on the basis, a data-driven learning algorithm is introduced, a TRP occurrence and rainstorm intensity grading recognition model is obtained through training, a risk assessment product is generated through calibration and risk grading, and therefore the inland typhoon long-distance rainstorm recognition and assessment capacity is improved.

The invention provides a rainstorm flood disaster chain monitoring field knowledge graph construction method, and relates to the technical field of knowledge graph construction, and the method comprises the steps: carrying out the preprocessing of original multi-source disaster data, and obtaining various types of quality-controlled data; calling a disaster field fine-tuning large model and an ontology inference engine based on the preliminary corpus and the multi-class quality-controlled data to generate a core ontology and a sub-document set, screening the sub-document set to construct a document association graph, and calling a mapping model to match a structured data field and a core ontology attribute; performing triple extraction according to the sub-document set and the constructed three-level cue word template; ontology instance attributes are vectorized and clustered, a three-layer collaborative system and an atlas core structure are generated through domain conflict repair and pruning density adjustment, a data calling disaster domain fine-tuning large model is obtained based on a processing process to complement an instance implicit relationship of the atlas core structure, and a target knowledge atlas is constructed. According to the invention, construction of the knowledge graph in the disaster field is realized.

The invention discloses a discrete element simulation method for slope unstable seepage, effectively solves the problem of nonlinear flow simulation distortion caused by excessive simplification of a traditional seepage model, and improves the calculation precision of transient processes such as rainstorm infiltration. The dynamic porosity feedback mechanism can reflect the influence of the internal structure change of the slope on the seepage field in real time, and more reliable pore water pressure prediction data is provided for landslide early warning. The establishment of a multi-scale coupling framework provides a new technical approach for unstable seepage analysis under complex geological conditions, a physical information neural network effectively avoids the overfitting risk of a pure data driven model, and an improved multi-objective optimization algorithm can quickly position an optimal parameter combination under complex constraint conditions. A dynamic updating mechanism can automatically adjust model parameters according to slope state changes, and the timeliness and accuracy of landslide early warning under the unstable seepage condition are remarkably improved.

The invention provides a vehicle management method based on visual identification. Comprising the following steps: performing visual image acquisition and basic feature identification on a vehicle in a target area, executing multi-source interference perception, acquiring raindrop volume concentration, a strong light incidence angle, vehicle tail gas heat flux density and a light intensity sudden change rate of the target area, and generating multi-factor coupling interference intensity data; executing dynamic exposure control based on the multi-factor coupling interference intensity data, outputting an exposure compensation coefficient and driving an image acquisition unit to adjust the shutter speed and the exposure duration; performing distortion correction feature extraction based on the image after exposure compensation coefficient correction, calculating image marginal definition and feature matching frequency deviation, and generating feature extraction confidence; judging whether to output a vehicle management recognition result or not according to the feature extraction confidence; the method can effectively deal with the composite interference scene, and solves the problems of low recognition precision and response lag.

The invention discloses an artificial intelligence rainstorm prediction model construction method based on multi-source heterogeneous data fusion, and relates to the technical field of rainstorm prediction, and the method comprises the steps: collecting multi-source meteorological data of a to-be-predicted region, the meteorological data comprising satellite remote sensing data, weather radar data, ground meteorological station data and numerical prediction mode data; dividing the to-be-predicted region into different dominant type regions based on a preset dominant factor discrimination rule, and constructing a virtual cloud cluster entity based on the dominant type regions; a motion model of a virtual cloud cluster entity is established, a life development track of the virtual cloud cluster entity is generated, and then a cloud cluster motion map reflecting dynamic evolution of the virtual cloud cluster entity is constructed; the system analyzes and generates an intervention instruction based on the cloud cluster motion map, projects the intervention instruction into the cloud cluster motion map, and finally outputs a rainstorm prediction report including deterministic early warning and potential risk assessment; according to the invention, a virtual cloud cluster entity system with dynamic perception and active intervention capabilities is constructed, and accurate prediction of the rainstorm formation process is realized.

The embodiment of the invention provides a disaster risk situation generation method and system based on dynamic grading, and a medium, and relates to the technical field of multi-disaster chain type disaster dynamic early warning. Comprising the following steps: performing rainstorm flood peak flow deduction by taking an infiltration loss parameter array as a constraint condition and combining an instantaneous rainstorm peak time-space field and a persistent rainfall characteristic time-space field to generate a disaster-inducing potential curved surface array; performing grid-level disaster susceptibility evaluation on the area static factors of the disaster monitoring area and outputting a reference susceptibility array; and constructing a dynamic grading-probability coupling array by coupling the disaster-causing potential curved surface array and the reference susceptibility array, executing disaster risk situation prediction based on real-time rainfall data, and outputting disaster chain risk probability distribution. The problems that in the prior art, a static single disaster type early warning system is adopted for split type static early warning, rainstorm-mountain torrent-geology chain type dynamic conduction risks cannot be captured, the dynamic conduction process of chain disaster is difficult to quantify, disaster chain response lags behind, and precision is insufficient are solved.

The invention discloses a data processing method applied to digital twinborn hydraulic engineering, and relates to the technical field of hydraulic engineering, the decision reliability of the digital twinborn hydraulic engineering in a rainstorm scene is significantly improved through algorithm architecture reconstruction, a two-channel cooperation mechanism deeply couples water level time sequence monitoring and rainfall space analysis, and the accuracy of data processing is improved. A graph convolutional network is used for extracting drainage basin level rainfall thermodynamic diagram features, self-attention dynamic noise filtering is combined, space-time alignment of meteorological data and ground sensing is achieved, false alarms caused by equipment interference or data asynchronization are restrained from the source, it is ensured that an early warning instruction is only triggered under the real flood situation condition, and the early warning efficiency is improved. The edge intelligent closed-loop decision breaks through the cloud dependence bottleneck, the lightweight model locally generates a scheduling strategy at a gate station node, the reinforcement learning driven gate control takes the water level deviation as an optimization target, and the loss of frequent actions on equipment is synchronously constrained, so that the flood discharge response is changed from passive rule execution to active risk stabilization.

The invention relates to the technical field of urban hydrological simulation and flood forecasting, and particularly discloses a comprehensive unit line improvement method applicable to an urbanized region, which comprises the following steps: S1, constructing an urban rainstorm runoff model, and simulating the flow process of a drainage basin outlet section under rainstorm in different recurrence periods; s2, calculating the flow process of the drainage basin outlet section under the same design rainstorm, and reconstructing and converting the flow process into a centralized flow process; s3, distributing the centralized flow process into sub-flows of a water outlet and a river channel section; s4, on the basis of the actual maximum drainage capacity of the drainage port, storage and drainage response correction is carried out on the sub-flow process of the drainage port, and a time sequence flow process considering drainage capacity constraint is formed; s5, calculating the segmented delay of the propagation time lag to the corrected flow, and obtaining an improved total flow process; the improved urban rainstorm basin outlet flow simulation is realized by performing space-time reconstruction, space distribution, drainage capacity constraint correction and propagation time delay delay on the flow process.

The invention discloses a progressive forecasting method for regional persistent rainstorm based on key impact factors, which comprises the following steps of: firstly, comparing and analyzing difference characteristics of a regional persistent rainstorm process and a general rainstorm process in a traditional physical quantity factor, a physical quantity anomaly factor and a comprehensive dynamic factor, and screening a key impact factor set; then, extracting a circulation field anomaly factor before rainstorm occurrence to construct a quantitative index system, and establishing a three-level linkage criterion in combination with a mid-high weft circulation form, a mid-low weft power lifting threshold value and water vapor conveying strength; and finally, determining an optimal factor combination and weight distribution scheme by taking the TS score as a judgment criterion, constructing a probability forecasting model by fusing a kernel density estimation technology, and outputting a rainstorm falling area spatial distribution probability. According to the method, a traditional single forecasting mode is broken through, a progressive forecasting technology system from rainstorm process forecasting to falling area probability forecasting is innovatively established, and technical support is provided for early warning decision making and disaster prevention and reduction deployment of extreme weather events.

The invention discloses a rainstorm flood risk early warning method and system based on machine learning and multi-source data fusion, and belongs to the technical field of flood prediction.The method comprises the steps that multi-source sample data is constructed based on flood disaster historical data; the multi-source sample data type comprises rainfall characteristic data, hydrological characteristic data, landform characteristic data, earth surface attribute characteristic data and social economic characteristic data; an XGBoost ensemble learning algorithm is adopted to construct a risk prediction model, and training and evaluation are carried out based on multi-source sample data; historical rainstorm flood event records are taken as labels in the training process; performing quantitative and application verification on the risk prediction model, and calibrating a risk level probability output by the risk prediction model based on a risk level distribution probability of historical disaster situation data; performing real-time estimation based on the optimized risk estimation model; and generating a spatial refined rainstorm flood risk grade early warning map in a future preset time period according to an estimation result in a rolling manner.