310 results about "Natural disaster" patented technology

The invention belongs to the technical field of emergency management, and discloses a natural disaster emergency rescue system based on multi-source sensing information fusion. The system is composed of a multi-source sensing data acquisition module, a data preprocessing and space-time registration module, a cross-modal feature extraction module, a multi-modal information fusion and conflict resolution module, a disaster type identification and grade discrimination module, a disaster influence range prediction and diffusion modeling module, and a dynamic emergency path planning and response plan generation module. A rescue scheduling and command control module; and an emergency feedback and closed loop dynamic correction module. Through multi-source sensing data fusion, cross-modal feature extraction and deep information fusion technologies, a full-space-time and full-process natural disaster emergency rescue system is constructed, comprehensive sensing, accurate recognition and dynamic plan generation of a disaster site are realized, the rescue response speed and decision scientificity are remarkably improved, and the intelligent level of emergency rescue is comprehensively improved.

The invention provides a disaster chain prediction method and system based on infrastructure cascade failure, and the method comprises the steps: S1, carrying out the screening and concrete of extreme natural disasters, and constructing an extreme natural disaster scene which accords with the characteristics of a research region; s2, constructing a single-layer network topology structure of various infrastructures and a cascade failure mechanism based on coupling characteristics of various infrastructures, and constructing an infrastructure multi-layer coupling network conforming to spatial distribution characteristics and actual operation logic of a research area based on the single-layer network topology structure and the cascade failure mechanism; and S3, superposing the constructed extreme natural disaster scene on the infrastructure multi-layer coupling network, and predicting a disaster chain spatio-temporal evolution process under the impact of the extreme natural disaster by taking the infrastructure multi-layer coupling network as a disaster propagation carrier. According to the invention, the disaster chain can be predicted.

The invention provides a hydropower station natural disaster intelligent early warning and emergency decision-making method and system based on multi-source data fusion and dynamic threshold optimization, and the method comprises the steps: collecting the water regimen data of a hydropower station and a drainage basin in real time through the deployment of a water level gauge, a rain gauge, a wind speed sensor and a thunder and lightning monitoring device, and combining weather forecast data and historical hydrological data, thereby achieving the intelligent early warning and emergency decision-making of the hydropower station. And dynamically calculating a disaster threshold value, and generating a graded early warning signal. The system comprises a data acquisition module, a dynamic threshold matching module, an early warning generation module, a case retrieval module, an information issuing module and a response tracking module. The early warning threshold is dynamically adjusted through multi-source data fusion and a machine learning technology, and the early warning accuracy is improved; a historical case library is constructed by using a knowledge graph technology, and intelligent recommendation of a disaster disposal scheme is realized; and early warning information and disposal suggestions are directionally pushed through multiple channels, so that the emergency response efficiency is improved. The conversion from passive response to active early warning is realized, and the influence of natural disasters on the operation of the hydropower station is obviously reduced.

The invention discloses an ice lake feedback monitoring and early warning method based on artificial intelligence and multi-source data fusion, and belongs to the technical field of natural disaster monitoring, and the method specifically comprises the steps: extracting ice lake state feature vectors of multiple spatio-temporal data through cross-modal self-supervised learning, and constructing a three-dimensional virtual ice lake model; based on the feature vectors, constructing a glacier three-dimensional stress field model by using a finite element method, constructing a seepage channel graph model by using a graph neural network, and fusing the features of the glacier three-dimensional stress field model and the seepage channel graph model to form mechanical-seepage coupling state vectors; constructing a pulse neural network to simulate glacier fracture extension and seepage mutation pulse events; constructing a heterogeneous graph federated architecture and outputting a federated weight matrix in combination with a dynamic weighted aggregation strategy; the federal learning weight and the three-dimensional model output result are fused, the outburst posterior probability is calculated through a Bayesian neural network, and a fourth-level early warning signal is generated; according to the invention, the multi-dimensional characterization and coupling process simulation of the state of the ice lake is realized, and the early warning accuracy and timeliness are improved.

The invention discloses a group drought and flood sudden change event identification method and system, and relates to the technical field of natural disaster risk assessment, and the method comprises the following steps: S1, collecting day-by-day rainfall data of a certain time period; s2, calculating an early-stage rainfall index (IAP) based on the day-by-day rainfall data, and standardizing the early-stage rainfall index (IAP) to obtain a standardized early-stage rainfall index (IsAP); s3, based on a run-length theory, respectively screening out a drought event and a flood event; s4, identifying a single-station drought and flood sudden change event, and calculating drought and flood sudden change intensity and sudden change point parameters; and S5, on the basis of the single-station drought and flood sudden change event, identifying a group-occurring drought and flood sudden change event, defining the comprehensive strength of the group-occurring drought and flood sudden change event, and performing grade judgment. The method is clear in physical significance, high in operability and capable of accurately recognizing the group drought and flood sudden turning events.

The invention discloses a distribution network-microgrid toughness improvement robust planning method and device considering natural disasters, and the method comprises the steps: constructing operation models of a distribution network and a microgrid under the natural disasters respectively, and carrying out the integration to obtain a distribution network-microgrid collaborative optimization operation model considering the influence of the natural disasters; generating a random interruption scene of a distribution line, a micro-grid line and a traffic road under the natural disaster; searching a scene which enables the operation cost of the power distribution network to be maximum, and obtaining a most serious natural disaster scene; considering that the distribution network-micro-grid encounters the most serious natural disaster, a distribution network-micro-grid toughness improvement robust planning model is constructed by taking minimization of the total investment cost of distribution network planning as a target, and an optimal toughness improvement scheme is solved. According to the method, on the premise of considering economy, a pre-disaster toughness improvement scheme can be planned for the distribution network-microgrid, and the load shedding amount of the distribution network under natural disasters is effectively reduced.

The invention discloses a landslide disaster risk dynamic assessment method and system based on an improvement project progress, and relates to the technical field of natural disaster treatment. According to the invention, through integration of multi-source heterogeneous data, real-time dynamic assessment and a machine learning model, accurate and dynamic assessment of landslide disaster risks is realized, assessment precision and stability are improved, and timeliness and pertinence of decision making are enhanced; a feature disturbance stability mechanism is introduced into the SPY algorithm, and a pseudo-negative sample screening strategy with higher robustness and discriminability is constructed; an information entropy reduction amount is introduced, an acquisition function is improved, the efficiency of hyper-parameter optimization is improved, and a stacking generalization integration model is effectively trained; the influence of the regulation situation progress in the landslide disaster area is considered, the susceptibility correction coefficient is set, the landslide disaster susceptibility evaluation result is corrected, it is ensured that the landslide disaster susceptibility evaluation result can reflect the actual situation in time, and powerful technical support is provided for prevention and management of landslide disasters.

The invention discloses a chaos annealing ant colony-based adversarial blocking timeliness emergency material transportation toughness intelligent decision-making method, and relates to the technical field of intelligent emergency scheduling and optimization. The method comprises the steps of 1, extracting scene data and converting the scene data into model parameters; step 2, constructing a mathematical model for survivability optimization; 3, initializing algorithm parameters and generating an initial path; 4, dynamically adjusting the temperature and the pheromone volatilization rate; 5, generating an adversarial blocking perception path and implementing detour repair; 6, updating pheromone distribution based on an annealing criterion; and step 7, locking the critical path and outputting an optimal transportation scheme. According to the method, timely delivery of materials can be guaranteed in a complex road network blocking and emergency demand scene, the timeliness and toughness of an emergency transportation system are improved through a chaos annealing mechanism and an adversarial blocking simulation technology, the method can be widely applied to the fields of natural disaster rescue, public health event response and the like, and a scientific basis is provided for emergency decision making.

The invention provides an emergency rescue path planning method and system for adaptive risk avoidance. The method comprises the following steps: establishing an emergency rescue road network structure; the method comprises the following steps: before an actual natural disaster comes, carrying out risk entropy driven adaptive Monte Carlo simulation analysis, carrying out path information statistical analysis on a Monte Carlo simulation result, and establishing a pre-planning path library; when an actual natural disaster comes, prior risk distribution of each planned path in the pre-planned path library is established; performing dynamic correction on the prior risk distribution of each planned path in the emergency rescue process by adopting sequential Bayesian reasoning; and carrying out emergency rescue path dynamic planning based on the path posterior risk mean value of each planning path after sequential Bayesian reasoning, selecting an optimal rescue path, and executing an emergency rescue task until the emergency rescue activity is finished. According to the invention, through a path dynamic adjustment strategy, a decentralized transportation strategy, a path pre-planning strategy and a network-free communication strategy, rescue resources are ensured to efficiently and safely arrive at a target location in natural disaster scenes such as flood and typhoon.

The invention discloses a mountain tourism route planning method based on weather and natural disasters, and the method comprises the steps: 1, building a Markov game model based on POMDP, combining Mama multi-modal feature compression and RAG semantic enhancement, designing a dynamic risk matrix integrating geological disaster probability, weather trend and ecological sensitivity, and providing partial observable environment state perception; 2, establishing a Nash equilibrium constraint optimization model, sharing an extreme weather strategy through federal learning, introducing an NSGA-II-Actor-Critic architecture to optimize unmanned aerial vehicle inspection and path adjustment, and triggering dynamic communication when rainfall exceeds a threshold value; 3, fusing a security and experience income design reward function, processing continuous action planning by using improved DDPG, and accelerating multi-target Pareto frontier convergence in combination with NSGA-II; 4, constructing a user weather sensitivity portrait, and dynamically adjusting weight balance income and AR simulation preview preference; and 5, based on a GIS closed-loop decision engine, performing agent and environment feature complementation, recording an emergency log by a block chain, and constructing a disaster case library by meta-learning.

A deformation monitoring device for geological landslide relates to the technical field of natural disaster monitoring and early warning, and comprises two monitoring pier bodies, a pull rope, a fixed seat, a support, a stay wire mechanism and an energy storage mechanism, the two monitoring pier bodies are correspondingly arranged on two sides of a ground crack, the pull rope is arranged between the two monitoring pier bodies in a tightened state, and the fixed seat is arranged on the support. The number of the fixing seat is one, the fixing seat is fixedly arranged between the two monitoring pier bodies in the front-back linear direction, the support is installed on the upper surface of the fixing seat, and the pull rope mechanism and the energy storage mechanism are arranged in the middle of the pull rope, so that the problem that the pull rope is interfered by external sundries is effectively solved; the motor drives the screw rod threaded block to reciprocate front and back, the wire hooking groove in the connecting plate pulls the pull rope to the edge of the support and loosens the pull rope, the check block of the energy storage mechanism blocks the rebounded pull rope to store energy, and when the threaded block drives the connecting plate to reset, the energy storage pull rope is released, and sundries are shaken off through resilience force.

The invention discloses a sudden natural disaster fault scene generation and influence identification method and system, and belongs to the technical field of power system disaster risk assessment and toughness modeling, and the method comprises the steps: building an interdependence relation matrix based on a node set of a three-network coupling network model; calculating the damage probability of each node in combination with the geographic coordinates of the nodes, and generating an initial fault set by using a Monte Carlo sampling method; defining a node state vector, and performing fault cascade propagation simulation until the node state vector reaches a stable state; respectively identifying a power failure area, a communication out-of-control area and a traffic blocking area; and outputting the initial fault set, the stable state and the three regions together as a simulated disaster scene sample to finish identification. According to the method, the problem that multi-network cascade fault propagation and system interruption identification are difficult under the influence of complex disasters is solved, automatic generation of samples and accurate identification of disaster influence areas are supported, data support is provided for follow-up, and the method can be used for evaluation and resource scheduling of multi-network integration.

The invention relates to the technical field of natural disaster risk assessment, in particular to a multi-disaster disaster-inducing factor internal correlation analysis method, which comprises four steps of data preprocessing, Bayesian network correlation model construction, space-time dynamic evolution analysis and risk early warning threshold model construction. And deep coupling analysis of disaster-inducing factors of multiple disasters such as mountain torrent-debris flow and the like is realized. The method comprises the following steps: firstly, collecting 17 disaster-inducing factor data such as terrain and rainfall, performing dimensionality reduction through kernel principal component analysis, and calculating a dynamic weight by adopting an entropy evaluation method; then constructing a Bayesian network model to quantify conditional probability association; introducing a space-time attention mechanism to optimize a multi-scale coupling weight; and finally, establishing a coupling risk early warning threshold value and outputting a relevance intensity matrix. According to the method, the problem that a traditional single-disaster analysis method cannot capture the space-time linkage effect of disaster-inducing factors is solved, the risk early warning accuracy of multiple disasters is improved, and scientific decision support is provided for regional disaster prevention and reduction.

The invention discloses a landslide risk early warning method, device and equipment based on multi-source data fusion, relates to the technical field of natural disaster monitoring, and aims to solve the problem of low precision and timeliness of landslide risk early warning. The method comprises the following steps: collecting multi-source monitoring data of a target area in real time and constructing a multi-source parameter set, wherein the multi-source parameter set comprises rainfall characteristic data, soil hydrogeological parameters and slope morphological parameters; performing space-time discretization moisture migration simulation based on the multi-source parameter set, and generating space-time distribution data reflecting the dynamic change of the slope moisture field; calculating a slope stability index sequence according to the spatial and temporal distribution data to construct a prediction model reflecting a risk evolution trend; and setting a dynamic grading early warning threshold based on a prediction model output result, and triggering multi-grade early warning information when the risk grade reaches a preset condition.

The invention relates to the technical field of data acquisition and analysis, in particular to a natural disaster engineering rescue data demand analysis and acquisition method, which comprises the following steps: based on different natural disaster site information, identifying a current engineering rescue task type, constructing a mapping model between the task type and a data demand, and outputting a task feature vector; outputting a data demand priority matrix in combination with the disaster emergency level, the task timeliness requirement and the core component distribution; inputting the data demand priority matrix into an acquisition path planning module, and generating an acquisition task sequence in combination with field reachability analysis and sensing resource states; and executing a structured data acquisition operation according to the guidance of an acquisition task sequence, outputting a feedback correction factor to update a task feature vector, and realizing self-adaptive iteration of a data acquisition scheme. The method has the advantages of high adaptability, high precision, high resource utilization rate and the like, and is suitable for natural disaster engineering rescue scenes with variable task types, complex environments and limited decision windows.

The invention discloses an InSAR (Interferometric Synthetic Aperture Radar) and graph neural network combined earth surface deformation space-time prediction method, which is suitable for space-time prediction of earth surface deformation. Based on DS-InSAR, obtaining radar sight line-to-time sequence earth surface deformation of the high-coherence measuring points; calculating the geographic distance between the measuring points and the mutual information between the deformation sequences corresponding to the points; the method comprises the following steps: judging connectivity between high-coherence measuring points according to a geographic distance between the measuring points and mutual information to generate an adjacent matrix, and organizing an original InSAR deformation monitoring result into graph structure data; and inputting the graph structure data into the LSTM-GCN model to predict and obtain the surface deformation of all the high-coherence measuring points in the whole area. The method is high in prediction precision and wide in application range, and can be effectively applied to the fields of space-time prediction of deformation of earth surfaces and buildings (structures) caused by mine closure, underground resource development, natural disasters and the like.

The invention relates to the technical field of grain storage, and discloses a novel heat preservation silo which comprises a silo wall, and the silo wall serves as a foundation of the whole silo and is used for bearing and building other structural parts. The granary top is located at the top of the granary wall, the bottom of the granary top is attached to the top of the granary wall, the granary top is used in cooperation with the granary wall, the granary wall comprises a wall plate, a side face heat preservation layer and a side face lining layer, the side face heat preservation layer is arranged on the inner side of the wall plate, the side face lining layer is arranged on the inner side of the side face heat preservation layer, the wall plate is made of galvanized steel, and the side face lining layer is arranged on the inner side of the side face heat preservation layer. The side face heat preservation layer is made of polyurethane foam, and the side face lining layer is made of reinforced concrete. By adopting the innovative structure that the outer-layer steel plates are matched with the inner-layer reinforced concrete, the design of the barrel-shaped wall plates and the vault accords with the mechanics principle, stress is dispersed, large loads can be borne, natural disasters such as earthquakes and typhoons can be effectively resisted, and compared with a traditional steel silo, the firmness is greatly improved.

The invention belongs to the field of natural disaster monitoring and early warning information processing, and discloses a landslide monitoring system and method based on multi-source data fusion. A data preprocessing module; a data processing and analyzing module; a dynamic early warning module and a remote visualization module; the method comprises the following steps: S1, installing equipment; s2, collecting data; s3, data preprocessing; s4, data fusion analysis; s5, risk judgment and early warning; s6, displaying a result; by adopting the technical scheme, the environmental factors can be monitored in real time, the environmental factors are fused into the basis of early warning judgment, the influence of the induction factors on the stability of the mountain can be sensed in advance, the comprehensiveness of landslide judgment is improved, and the accuracy and sensitivity of mountain early warning are improved; meanwhile, according to the technical scheme, the threshold value is optimized, so that the system judges whether the threshold value is a fixed threshold value or not, the threshold value can be dynamically adjusted according to the monitored environmental factors, and the early warning of the system on the landslide is more accurate.

The invention discloses a temporary emergency shelter site selection method and system suitable for a huge disaster scene, and relates to the technical field of natural disaster emergency management. According to the method, disaster data and remote sensing image data of a disaster area are processed, disaster grids and non-disaster grids in the disaster area are identified, the site selection range is narrowed, and the site selection precision and accuracy are further improved; risk evaluation, openness and traffic evaluation are carried out on the non-disaster grids, based on the evaluation result, the grey wolf algorithm is utilized to carry out site selection on the non-disaster grids, a temporary emergency shelter site selection scheme is obtained, through data driving, manual analysis dependence is reduced, the site selection automation degree is improved, the site investigation requirement is reduced, and efficiency is improved. And through integration and rapid analysis of multi-source data, dynamic response to change demands is realized, dynamics is enhanced, and a site selection strategy is updated through real-time data, so that emergency service coverage is ensured to be matched with actual demands.

The invention provides a model workflow-based natural disaster landslide early warning risk assessment method and system. The method comprises the following steps: sorting index factors and algorithm models according to different scales; constructing a tool set and issuing the tool set; according to risk assessment requirements, calling the index factor tool set and the model tool set, and constructing a natural disaster landslide early warning risk assessment workflow; calling the natural disaster landslide early warning risk assessment workflow to carry out risk assessment to obtain a model result data set; and based on the model result data set, in combination with different matching colors corresponding to different risk values, issuing wms map service. According to the invention, different landslide disaster models can be flexibly constructed, and the protection work of natural disaster landslide can be well done.

The invention discloses a rainstorm disaster loss prediction method and system and a medium, and belongs to the technical field of natural disaster loss prediction. Comprising the following steps: acquiring disaster-inducing factors causing rainstorm disasters, disaster-pregnancy environmental factors breeding the rainstorm disasters and factors reflecting exposure and vulnerability of disaster-bearing bodies in the rainstorm disasters; through Min-Max standardization, scale differences of disaster-inducing factors, disaster-pregnancy environment factors and exposure and vulnerability factors are eliminated; adjusting the sequence of the disaster-inducing factors, the disaster-pregnancy environment factors and the exposure and vulnerability factors by using a random sampling mode; integrating a plurality of machine learning algorithms after hyper-parameter combination optimization, and constructing a multi-model rainstorm disaster loss prediction system; inputting the processed disaster-inducing factors, disaster-pregnancy environment factors and exposure and vulnerability factors into a multi-model prediction system, and outputting economic loss of the rainstorm disaster; according to the method, the economic loss of the rainstorm disaster is effectively predicted.

The invention relates to a natural disaster early warning and response method based on constrained multi-task optimization, and belongs to the technical field of natural disaster early warning and resource scheduling, and the method comprises the steps: obtaining a task list containing multiple disaster prediction, resource scheduling, path planning and rescue cooperation tasks, building a population based on the multiple tasks, and carrying out the prediction of the multiple disaster prediction, resource scheduling, path planning and rescue cooperation. A multi-factor evolutionary algorithm is utilized to optimize a plurality of individuals in the population with the minimum multi-disaster joint prediction error, the maximum material coverage rate, the minimum transportation time, the minimum path, the maximum wounded rescue rate and the minimum response time, and an optimal individual and a corresponding response scheme thereof are obtained; the scheme covers a Pareto optimal solution set, an optimal resource scheduling strategy, an optimal path planning strategy and an optimal rescue cooperation strategy. The method has the technical effects that multiple tasks can be efficiently coordinated, and complex disaster scenes can be dynamically adapted.

The invention provides a disaster response method, device and system and control equipment, relates to the technical field of natural disaster early warning, and is used for helping personnel at a disaster occurrence place to know disaster early warning information in time on the basis of guaranteeing the authenticity of disaster early warning. The method is applied to a control device in the disaster response system, the disaster response system further comprises an energy storage street lamp connected with the control device, and the energy storage street lamp is provided with a plurality of sensors. The method comprises the following steps: a control device obtains authoritative early warning data, and determines a disaster type and a disaster occurrence place indicated by the authoritative early warning data; and acquiring local sensing data acquired by a target sensor on the target energy storage street lamp. Further, the control equipment verifies whether a target disaster exists in the disaster occurrence place or not according to the authoritative early warning data and the local sensing data; and controlling the target energy storage street lamp to execute a disaster response strategy corresponding to the target disaster under the condition that the target disaster exists in the disaster occurrence place.

We provide the system. [Solution] A method that uses generative models to predict future natural disasters by analyzing environmental data collected from diverse sources, A means for providing an appropriate evacuation route to a user terminal based on the predictions of the generation model, A means of sharing community information among users in real time and promoting its use in disaster prevention, A method for using unmanned aerial vehicles to acquire information on the situation in disaster-stricken areas and then re-proposing the optimal evacuation route based on that information, A system that includes this.

The invention discloses a dynamic flood risk evaluation method based on flood inundation simulation, and relates to the technical field of natural disasters. The method includes: acquiring meteorological and hydrological data of a target area; constructing a flood risk model according to the meteorological and hydrological data, and carrying out flood risk model parameter calibration by utilizing actually measured rainfall data and flow-water level data; based on a D8 diffusion algorithm and water balance equivalent analysis of a GIS spatial analysis technology, converting the water amount, higher than the river embankment, of the target area into a submerged water depth sum by using the constructed flood risk model, obtaining a submerged water depth map of the target area, and generating a flood deduction result; and according to the flood deduction result, determining flood disaster risk grade division of the target area, and evaluating the flood disaster risk. According to the method, the flood risk model and the flood rehearsal model of GIS spatial analysis are constructed, and the flood risk model can well reflect the flood rehearsal result.

The invention discloses a natural disaster early-warning method based on a low-orbit satellite, and the method comprises the steps: obtaining sensing data and satellite data of the low-orbit satellite, compressing the sensing data, carrying out the hierarchical transmission of the compressed sensing data, carrying out the edge calculation of the data transmitted in a hierarchical manner, obtaining key features, and carrying out the early-warning of natural disasters based on the key features. According to the key features, sensing data transmitted in an optimized mode are obtained; performing fusion analysis on the satellite data and the optimally transmitted sensing data to generate fusion data; natural disaster prediction is carried out according to the fusion data, and a natural disaster prediction result is obtained; performing trend prediction on the fused data, and performing judgment according to a trend prediction result to obtain a predicted natural disaster prediction result; historical disaster data are acquired, the sensing data acquisition position is optimized according to the historical disaster data, and the sensing data are acquired according to the optimized position.

The invention relates to the technical field of data analysis, in particular to a natural disaster engineering rescue data platform construction and intelligent analysis system, which comprises an edge intelligent sensing module for generating a field three-dimensional live-action grid through laser radar point cloud and visible light video fusion and extracting disaster feature vectors; the cloud deep analysis module adopts a graph neural network to aggregate multi-source data, and generates a global disaster evolution graph composed of disaster grade distribution and infrastructure vulnerability indexes; the dynamic decision generation module is used for generating a space-time cooperative rescue instruction set composed of an equipment scheduling path and an engineering disposal sequence in combination with a rescue resource library and engineering constraints including an equipment mechanical arm operation radius and an aftershock influence coefficient; and the closed-loop optimization module is used for collecting operation feedback logs, correcting extraction weights of disaster feature vectors through a difference comparison engine, and dynamically updating an engineering constraint library. According to the invention, unified cooperation of disaster sensing intelligence, task allocation optimization and response process closed loop is realized.

The invention relates to the technical field of electric power, and discloses an electric power natural disaster emergency resource allocation method and system based on multi-data fusion, and the key points of the technical scheme are as follows: multi-source data acquisition and fusion, resource demand prediction, dynamic allocation scheme generation, and real-time monitoring and adjustment. By constructing a multi-source data fusion system, establishing a high-precision demand prediction model and dynamically generating an optimal allocation scheme in combination with traffic accessibility, accurate and efficient overall planning of emergency resources is realized, the resource utilization rate is improved, and the power recovery time after disasters is shortened.

Techniques for calculating or determining risk scores for certain natural disasters perils based on machine learning model outputs are discussed herein. For example, a machine learning model may weight each of the pixels of a map in accordance with the set of weights associated with a structure, to calculate a risk score for a particular natural disaster peril associated with that structure. A plurality of risk selections may be provided to a user computing device for selection by a user, with those risk selections being associated with that risk score. Advantageously, the computing system facilitates the interaction of datasets with different measurement parameters in a machine learning model. In normalizing datasets before providing the datasets to input nodes of a machine learning model, a computing system may efficiently provide hazard and vulnerability outputs of the machine learning model.

The invention provides a multi-source data fusion flood risk assessment early warning method, and relates to the technical field of natural disaster early warning. The multi-source data fusion flood risk assessment early warning method comprises the following steps: data acquisition: acquiring meteorological data including but not limited to rainfall, air temperature, air pressure and wind speed; hydrological data including river flow, water level and soil humidity are collected; collecting geographic data including terrain elevation, land utilization type and vegetation coverage; social data including population density, building distribution and infrastructure conditions are collected; and data preprocessing: cleaning the collected data, removing noise and abnormal values, and carrying out standardization processing on the data to ensure that the data of different data sources has comparability. According to the flood risk assessment method, multi-dimensional features of flood risks can be comprehensively reflected by integrating multi-source data of meteorology, hydrology, geography, society and the like, and richer information support is provided by integrating different types of data, so that risk assessment is more comprehensive and accurate.