720 results about "Difference analysis" patented technology

The invention relates to the technical field of nondestructive testing, and discloses a steel structure weld defect ultrasonic detection and identification system. A data acquisition module of the system acquires an original ultrasonic signal of a steel structure welding seam through ultrasonic detection equipment and acquires geometric attribute data of the welding seam; the model construction module constructs a welding seam three-dimensional digital model based on the data; a feature extraction module performs feature mining on the three-dimensional digital model and extracts a weld defect feature index set; the difference analysis module carries out deviation calculation on the characteristic index set and a reference index set in a standard welding seam characteristic database, and an abnormal area is identified; the risk assessment module calculates a defect sensitivity index according to the abnormal region in combination with real-time environmental parameters, and assesses a defect risk level; and the report generation module formulates a detection scheme according to the defect risk level, generates a detection instruction, executes ultrasonic scanning, collects performance data and generates a defect detection report. The system has the advantages of high detection precision, high reliability, automatic and standardized process and the like.

The invention relates to the technical field of video behavior recognition, and discloses a behavior recognition and early warning system for a real-time video stream. The system comprises a spatio-temporal feature modeling module, a behavior fragment extraction module, an anomaly propagation modeling module, a risk area positioning module and an early warning strategy generation module. The spatial-temporal feature modeling module builds a dynamic model based on historical data, captures a skeleton key point three-dimensional coordinate sequence, a motion optical flow vector field and a micro-expression intensity spectrum, and outputs a theoretical behavior mode vector; the behavior fragment extraction module generates a multi-modal difference feature tensor through cross-modal difference analysis; the exception propagation modeling module generates an exception propagation path risk probability distribution cloud picture in combination with spatial constraint and trajectory information; the risk area positioning module identifies a high-risk area and marks a boundary; and the early warning strategy generation module dynamically configures monitoring parameters, starts high-frame-rate micro-expression capture for a high-risk area, and performs a track disturbance test on an adjacent area.

The invention belongs to the technical field of dam safety early warning, provides a dam safety monitoring method based on digital twinning, and aims at solving the problems that an existing monitoring method is weak in prediction capability, lags in early warning and the like. The method comprises the following steps: deploying multiple types of sensors to collect multi-dimensional physical state data; establishing a digital twinborn model containing a multi-physics field coupling simulation sub-model and data driving correction based on the BIM, the Internet of Things and finite elements; inputting a historical data driving model to predict an output state trend sequence and a state data value at a certain moment; performing difference analysis to construct an error vector, and calculating and correcting a deviation index; and calculating a structure health score, and if a threshold value is exceeded, early warning. The method has the advantages of improving prediction capability, enhancing adaptability, realizing comprehensive evaluation and providing systematic advanced decision basis for safety management.

The invention relates to the technical field of low-altitude safety emergency, and discloses a low-altitude safety emergency system based on large model application. The system comprises a low-altitude situation awareness module which constructs a dynamic situation prediction model based on historical flight data, collects a low-altitude target track coordinate sequence, radar echoes and communication signal characteristics in real time, and outputs a theoretical security situation value. The multi-source data fusion analysis module performs three-dimensional difference analysis on theoretical and actually measured situation values, which relates to time domain trend deviation, space domain coverage offset and event sequence matching degree, and generates a target-level difference feature matrix; and the risk propagation path deduction module inputs the matrix into a geographic topology analysis network, generates a probability thermal distribution diagram of an abnormal risk propagation path in combination with airspace node communication parameters and a target position, and locates a risk influence area. And the intelligent disposal strategy generation module configures parameters according to the graph, starts high-frequency monitoring for a high-probability area, and performs communication disturbance test on an adjacent target.

The invention relates to the technical field of electric power monitoring analysis, and discloses an intelligent analysis system for electric power monitoring data based on a mutual inductor. The system comprises a mutual inductor data acquisition and structuring module which acquires a current waveform, a voltage waveform and a harmonic component in real time, and generates a standardized monitoring data unit through preprocessing, field mapping and association identification establishment; the theoretical monitoring value calculation module constructs a dynamic calculation model according to historical power data, and an input standard data unit outputs a theoretical value; the rule conformity verification module is used for matching the equipment type with the operation rule and generating single compliance judgment; a multi-dimensional difference analysis module compares a theoretical value with a measured value from time domain deviation, frequency domain deviation and waveform distortion, and generates an equipment-level difference coefficient matrix; the association network construction module is used for constructing a multi-monitoring-point association map according to the equipment identifier and the position information; and the anomaly positioning and strategy generation module combines the map, compliance judgment and a difference matrix, calculates a risk score, generates an anomaly probability distribution map, positions anomaly and matches a monitoring strategy.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses an analysis method, device and equipment based on multi-modal information and a medium, and the method comprises the steps: obtaining visual information, sound information and physiological index information of a target object; the multi-modal sensing information is preprocessed, and standard multi-modal information is generated; extracting facial characterization features, behavior characterization features, voice characterization features and physiological characterization features; generating a state representation of the target object based on the multi-dimensional features; comparing the state characterization with a preset reference standard to generate a state difference result; and generating an analysis result according to the state difference result. According to the method, the multi-modal perception information is fused, the multi-dimensional features are extracted, the state representation is constructed, and the difference analysis is carried out, so that the analysis result has the characteristics of quantifiability, comparability and the like, and the objectivity and accuracy of evaluation are improved.

The invention relates to a multi-stage filtering road spilled object detection method based on dynamic difference analysis, which is suitable for automatic identification of unstructured foreign matters in video monitoring. The method comprises the following steps: firstly, extracting a reference image road mask, eliminating vehicle and pedestrian interference by using YOLOv8 detection, and extracting a motion candidate area through a frame difference method and background modeling; and then context expansion and super-resolution reconstruction are carried out on the candidate region, the candidate region is converted into an HSV space, multi-dimensional features such as color similarity, structural similarity and shadow determination are synthesized for screening, false detection is further removed in combination with inter-frame time sequence consistency, and finally a stable detection result is output. The method provided by the invention has the advantages of strong anti-interference capability, high adaptability, high detection precision and the like, and is suitable for the intelligent recognition task of the expressway thrown objects in a complex environment.

The invention provides an interactive simulation teaching processing method and device for ship electromechanical guarantee, and relates to the technical field of interactive simulation teaching, and the method comprises the steps: building a ship electromechanical full-structure model in a virtual simulation platform, digitally mapping real equipment operation parameters into virtual parts with real-time response characteristics, and carrying out the real-time simulation of the virtual parts; synchronous acquisition and normalized calibration of operation instructions, equipment feedback and environment disturbance in a teaching scene are realized, and a unified teaching simulation data set is generated. State recognition, fault simulation and operation evaluation calculation are completed through multi-layer task scheduling, and real-time teaching feedback is generated and visually displayed on a terminal. Based on difference analysis of dynamic evaluation data and a standard operation template, operation deviation and abnormity are automatically identified, and a targeted guide prompt is generated. According to the invention, unified modeling and data linkage can be carried out on student operation behaviors, equipment dynamic responses and teaching feedback results, and the problems of teaching feedback lagging and lack of pertinence in operation correction are solved.

The invention relates to the technical field of earthwork compaction, and discloses a pipe ditch backfilling and tamping system based on excavator modification. The system comprises an earthwork condition dynamic sensing module which captures a bucket tooth track, a hydraulic oil cylinder pressure pulse and a soil mass vibration spectrum in real time and combines a historical database to construct a model to output a theoretical compaction degree curve; a multi-source compaction difference analysis module performs three-dimensional deviation detection on the theoretical curve and the actual settlement volume data to generate a construction grid-level compaction difference tensor; the defect area association mapping module inputs the tensor into a geological topology analysis network, generates a loose defect propagation path probability cloud chart by combining soil parameters and identifies weak sections; and the tamping strategy self-adaptive configuration module automatically sets operation parameters according to the cloud picture, activates high-precision monitoring on a high-probability defect section, and applies a dynamic excitation force test on adjacent grids, so that intelligent management and control of the backfilling and compacting process are realized.

The invention relates to the technical field of data processing, can be applied to the field of science and technology finance / digital medical treatment, and discloses a multi-source heterogeneous data verification and tracing method and device and related equipment. The method comprises the following steps: acquiring multi-source heterogeneous data from a service data source of a target service; preprocessing the multi-source heterogeneous data, extracting a first data field in each data source, and matching the first data fields with the same semantics in different data sources based on a preset semantic mapping rule; performing consistency verification on the matched first data fields to obtain second data fields with inconsistent verification results; performing root cause tracing on the second data field based on a preset difference analysis rule, and determining a difference type corresponding to the second data field according to a root cause tracing result; and generating a multi-source heterogeneous data verification and tracing result according to the business data source, the second data field, the root cause tracing result and the difference type. According to the method, matching, difference identification and root cause positioning of multi-source heterogeneous data are realized.

The invention relates to the technical field of power system automation, in particular to a monitoring and maintenance system and method for a transformer substation. The system comprises a quantitative classification module, a difference analysis module and a grading adjustment module. Equipment in a transformer substation is classified through the quantitative classification module, correction parameters such as aging and load loss are introduced for single-equipment monitoring through the difference analysis module, misjudgment and missed judgment caused by equipment state changes are reduced by combining multi-level threshold values, associated equipment is modeled through a GNN model, neighbor states are aggregated, the equipment health degree is output, and the equipment quality is improved. The method comprises the following steps of: quantifying a fault propagation probability, pre-warning cascading fault risks in advance, and adjusting a threshold value of a single device in a linkage manner while a hierarchical adjustment module takes risk response measures, so that single device monitoring is adaptively matched with an associated device risk state, full-coverage accurate monitoring is realized, and local and global risks are considered.

The invention discloses a live-action three-dimensional dynamic change modeling method based on digital twinning. The method comprises the following steps: collecting multi-source data; carrying out space registration and time synchronization processing, and outputting a fusion data set; carrying out three-dimensional modeling, optimizing model parameters, and generating a live-action three-dimensional model; virtual-real mapping is carried out, and generated virtual space data and state parameters thereof are stored; performing differential analysis to generate a change detection result data set; performing dynamic change modeling to generate an updated virtual model; executing geometric accuracy optimization and semantic consistency check, and outputting an optimized virtual model; bidirectional data flow association and dynamic updating are executed, and the live-action three-dimensional dynamic change modeling process is completed. The real scene dynamic updating is realized by adopting the improved Gaussian sputtering modeling, and the method has the advantages of high precision, strong real-time performance and virtual-real synchronization.

Provided in the present invention are a method and system for real-time update of a medical knowledge base, and a medium and a device. The method comprises: acquiring medical literature data in real time; analyzing the medical literature data, so as to obtain first structured data; performing correlation analysis on the first structured data and second structured data in a medical knowledge base, so as to obtain an analysis result; on the basis of the analysis result, determining whether to update the medical knowledge base; identifying the differences of the medical knowledge base before and after the update, so as to obtain a difference identification result; assigning a unique version number to each difference in the difference identification result, so as to obtain version update records; and on the basis of the version update records, using incremental hash table technology to perform data storage for each difference. The present application has a higher update efficiency without the risk of omission; and by means of difference analysis and by using incremental hash table technology to store difference data, the present application significantly reduces data storage space requirements, and also improves data access speed.

The invention relates to the technical field of atomic clock control, and discloses a punctuality type atomic clock control method based on a microwave frequency band. The method comprises the following steps: firstly, acquiring microwave frequency band signal data and atomic state data, and constructing a time calculation model in combination with historical operation data to output a theoretical time value; performing multi-dimensional difference analysis on the theoretical time value and the actual measurement time value to generate a time difference coefficient matrix; inputting the matrix into a spatial propagation analysis network, and combining atomic clock component parameters and environmental parameters to generate a time anomaly probability distribution diagram; and finally, configuring microwave control parameters according to the distribution diagram, including starting a high-frequency monitoring mode for a high-probability abnormal region and applying a disturbance test to adjacent frequency bands. According to the method, through multi-dimensional data modeling, difference analysis and dynamic parameter configuration, the accuracy and adaptability of atomic clock control are improved, and the operation stability of the atomic clock in a complex environment can be enhanced.

The invention relates to the field of concrete detection, and discloses a concrete structure health real-time monitoring system based on multi-source sensing fusion, and the system comprises a data collection unit which carries out the collection of the physical field data of a concrete structure through a sensor array, and obtains the multi-source sensing data; performing structure response difference analysis on the multi-source sensing data to obtain concrete structure response difference data; according to the response difference data of the concrete structure, performing spatial and temporal distribution density analysis on a microcrack propagation path to obtain microcrack propagation spatial and temporal distribution density data; a data evaluation unit; multi-scale damage evolution path simulation is carried out through the structural damage time-space correlation characteristic data to obtain multi-scale damage evolution path data, the sensor array is used for collecting multi-source sensing data, structural response difference analysis is carried out, tiny changes of a concrete structure can be accurately captured, potential structural problems can be found in time, and the construction efficiency is improved. And a high-quality data basis is provided for subsequent analysis.

The invention provides an application environment and version management method and system suitable for a low-code platform, and relates to the technical field of platform management.The application environment and version management method comprises the steps that a directed dependency graph is constructed by obtaining change records, and a dependency loop is detected and intelligently decomposed through a depth-first traversal algorithm; and packaging the node clusters into micro-services based on a function relevancy model to form an acyclic directed dependency graph, constructing a version backtracking index and a difference analysis matrix to determine a change influence range, executing incremental deployment according to a topological sequence, and recording deployment information. According to the method, the problem of version management and environment dependence in a low-code platform is solved, and the application deployment efficiency and the system stability are improved.

The invention relates to the technical field of laser cutting, and discloses a high-power laser cutting light path system, a laser cutting head and a using method of the laser cutting head. The system comprises a dynamic cutting path accounting module, a multi-dimensional cutting difference analysis module, an abnormal region association positioning module and a self-adaptive cutting strategy generation module. The dynamic cutting path accounting module builds a model based on laser cutting historical data, and collects laser power fluctuation, light beam quality distribution and cutting speed sequence data of current cutting in real time. The multi-dimensional cutting difference analysis module receives the information and performs three-dimensional difference analysis with an actual measurement path of the optical sensor to generate a workpiece-level difference coefficient matrix; and the abnormal region association positioning module inputs the matrix into a space geometric analysis network, generates an abnormal probability thermal distribution diagram of a cutting defect propagation path, and positions an abnormal cutting region. And the adaptive cutting strategy generation module configures parameters according to the thermodynamic diagram, and applies multi-band impedance disturbance to the adjacent node cutting head with the maximum abnormal probability gradient change.

The invention relates to the technical field of industrial Internet of Things equipment fault diagnosis, and discloses a power plant equipment fault diagnosis method and system based on the Internet of Things, and the method comprises the steps: obtaining an equipment working condition monitoring data set, executing coupling analysis and abnormal triggering collection, and obtaining a multi-equipment association state data set; performing node analysis and path distribution based on the data set to form a fault diffusion path scheme; carrying out risk calculation and trend reasoning to obtain a real-time fault risk assessment result; according to an evaluation result, searching an optimal parameter combination from a preset parameter library and generating a parameter instruction set; after the instruction set is executed, initial operation state data are collected, delay evaluation and parameter correction are carried out, and final operation state data are obtained; and performing difference analysis and risk scoring according to the state data, adjusting a path weight, and updating a fault diffusion path scheme. The method can solve the problem of fault diagnosis lag.

The invention relates to the technical field of sharing management, in particular to an instrument sharing management method and system based on a cloud platform, and the method comprises the following steps: obtaining a task generation structure sequence, generating and publishing screening equipment, constructing a channel list generation partition table, and binding a resource output management record. According to the method, the experimental task text content is automatically extracted to perform refined analysis on the behavior structure relationship, the cooperative relationship between the devices is accurately split according to the operation sequence and the behavior difference, and the device response capability and the resource occupancy characteristic are quantitatively identified through the multi-dimensional index; the conflict relation is executed to jointly judge by difference analysis and resource intersection, a priority execution path is constructed in the equipment operation sorting and channel scheduling process according to a response level, and task blocks are accurately divided by jointly deducing an execution proportion and a starting interval in operation time interval arrangement; and automatic closed-loop linkage management of task analysis, equipment identification, conflict detection, priority ordering, block division, resource allocation and sharing registration is realized.

The invention relates to an AI-based supply chain demand prediction and inventory optimization method and system, and the method comprises the following steps: calling a sales record from a supply chain, recognizing a periodic demand rule of a corresponding commodity, and predicting an estimated demand quantity in a future time period according to the periodic demand rule; carrying out supply and demand difference analysis by combining the current warehouse commodity stock margin, and judging the stock state; if the commodity stock surplus exists, making a corresponding stock digestion strategy; if a shortage quantification result appears, carrying out key degree evaluation and urgency sorting based on a shortage amount in combination with market factors, and generating a key commodity list and an out-of-stock urgency index; and finally, in combination with a preset commodity constraint condition, according to the key commodity list and the urgency index, an optimal replenishment scheme meeting resource and time limit is formulated, and the technical problem of how to realize conversion from passive response to active pre-judgment in a supply chain environment with large demand fluctuation and complex influence factors is solved.

The invention discloses a semantic tag generation method based on a power grid engineering corpus, particularly relates to the technical field of power grid system fault analysis, and is used for solving the problem that the existing semantic tag technology cannot accurately represent dynamic association and physical propagation paths of composite events. Equivalent traveling wave propagation delay of an event entity is extracted based on an electromagnetic transient waveform, and a sequential relation graph is constructed to restore a fault dynamic evolution process; establishing a physically interpretable causal chain hierarchical structure through equipment topology path matching and wave impedance gradient difference analysis; a gradient clustering algorithm is adopted to adaptively adjust label granularity confidence, noise interference is eliminated, and a key propagation path is reserved; and finally outputting a versioned label set fusing the dynamic causal chain, the traveling wave propagation time delay and the equipment physical association path, and forming a closed-loop modeling mechanism of event space-time association and equipment topology linkage.

The invention relates to the technical field of unmanned vehicle path planning, and discloses an unmanned vehicle dynamic path planning method based on multi-source sensor fusion. The method comprises the following steps: an environment state dynamic coding step: generating an environment state coding vector through a dynamic coding model; a prediction path generation step of generating a prediction path sequence of the unmanned vehicle; a multi-dimensional path difference analysis step: performing multi-dimensional difference comparison on the predicted path sequence and real-time environment data to generate a path difference coefficient tensor; a risk area probability positioning step: generating a risk probability distribution diagram through space environment topology analysis based on the path difference coefficient tensor; a dynamic path cost calculation step of calculating a dynamic path cost vector in combination with the risk probability distribution diagram and the environment state coding vector; a path planning strategy optimization step: optimizing a path planning strategy through a strategy learning model; and a real-time dynamic path generation step: generating a real-time dynamic path based on the optimized strategy.

The invention discloses a poisoning defense method based on characteristic difference analysis and model layer purification. The method comprises the following steps: S1, constructing a poisoning classification model according to an electromagnetic signal sample; adversarial disturbance is introduced, cross entropy loss of disturbed samples is calculated and sorted, and a threshold value is set to distinguish clean samples from poisoned samples; s2, after the poisoning samples are separated out, the poisoning score of each layer of the model is calculated through quantification, and the higher the score is, the stronger the influence of the back door neurons of the layer on the poisoning samples is; s3, generating a pseudo-poisoning sample based on an inversion trigger, analyzing the feature distribution difference between the pseudo-poisoning sample and a clean sample, aligning a feature space and purifying a model layer by optimizing an objective function, and enhancing the distinguishing ability of the model to the sample; s4, repeating the step S3 until the model classification precision converges, and storing the optimal network parameters; and the classification precision of the model on a normal sample and the attack success rate on a poisoning sample before and after defense are evaluated. According to the method, the robustness and the safety of the model are improved, and the method has relatively high universality.

The invention relates to the technical field of anti-collision systems, in particular to an offshore wind power construction safety pre-warning system based on AIS (automatic identification system) data, which comprises a ship dynamic uncertainty modeling module for acquiring AIS signal updating frequency and positioning precision marks of a target ship and generating a ship future position probability ellipse. According to the invention, the position coordinate, the speed value and the course angle of the ship are obtained in real time based on the AIS data, the reliability and the motion trend of ship position prediction are determined by combining the AIS signal updating frequency and the positioning precision mark, and the future position probability ellipse of the ship is accurately generated; a protection area is dynamically constructed, comprehensive calculation is carried out in combination with the spatial relation between a ship future position probability ellipse and the positions of surrounding fan pile foundations, and a hourly dynamic collision risk index sequence is obtained; and further according to the statistical deviation between the current motion state of the ship and the standard parameter of the function partition, combining with path difference analysis to obtain a navigation abnormity comprehensive index.

The invention discloses a method and system for detecting trace organic pollutants in a water body, and relates to the field of environmental water body detection.The method comprises the steps that environmental parameters of a target water body sample to be detected are collected, and baseline spectral data of the target water body sample are obtained; determining disturbance operation parameters matched with the target water body sample, and applying micro-disturbance to the target water body sample according to the disturbance operation parameters to obtain disturbance response spectral data; performing time sequence differential analysis on the disturbance response spectral data and the baseline spectral data to obtain a disturbance response residual error spectrum, and identifying an optimal response characteristic wavelength from a characteristic response wavelength interval of the target trace organic pollutants based on the disturbance response residual error spectrum; and solving the concentration correction model according to the optimal response characteristic wavelength, the disturbance operation parameters and the environmental parameters to obtain the concentration of the target trace organic pollutants. Therefore, the detection sensitivity and the anti-interference capability of trace organic pollutants in the water body sample are improved.

The invention relates to the field of agricultural data analysis, in particular to a straw returning effect analysis method combining soil nutrients and crop yield, and the method comprises the steps: collecting a multi-source feature data matrix for risk assessment; performing conjoint analysis on the crop yield data and the soil attribute data to obtain a static nitrogen deficiency risk amplification factor; performing difference analysis on the vegetation index data, the yield data and the cluster centroid to obtain a dynamic growth response asymmetric regulatory factor; carrying out fusion evaluation on the static nitrogen deficiency risk amplification factor and the dynamic growth response asymmetric regulation factor to obtain risk perception asymmetric distance measurement based on yield data; the clustering process of risk perception asymmetric distance measurement based on yield data is subjected to zoning optimization, and an optimal management zoning result for formulating a variable fertilization strategy is obtained, so that the nitrogen deficiency risk of crops in the seedling stage is reduced, and the nutrient utilization efficiency and the yield stability are improved.

The invention relates to the technical field of environment monitoring and early warning, in particular to a forest fire monitoring and early warning system based on big data, which performs multi-period data superposition calculation in combination with a temperature change difference value, remarkably improves the recognition precision of a suspected point of a fire source, enhances the capture capability of dynamic change of forest environment temperature, and improves the accuracy of forest fire monitoring and early warning. According to the method, the identification rate of potential fire behavior is improved, difference analysis is performed by using a time period highly overlapped with humidity in a temperature trend curve, a high-risk section is accurately marked in combination with the temperature rise rate and the wind direction continuity, the judgment capability of a fire risk range is improved, and the judgment is matched with the change rate of a risk point through path overlapping number judgment. Quantitative control of trigger conditions is realized, so that a risk control decision has multiple response capabilities of spatial dimension, time dimension and dynamic evolution characteristics, and timeliness, spatial accuracy and risk response perspectiveness of forest fire prediction are improved.

The invention discloses a forestry remote sensing image recognition method and system based on image recognition, and relates to the technical field of image processing, and the method comprises the steps: collecting and preprocessing a LiDAR point cloud and a multispectral optical image, and obtaining a three-dimensional forest point cloud; forest feature segmentation is carried out through spectral difference analysis, forest feature segmentation boundary points are redistributed based on spectral dissimilarity, a core spectral signature is calculated, and a single-tree point cloud entity set is output; and carrying out topological analysis on the single tree point cloud entity set, and outputting a vector result. According to the method, the three-dimensional forest point cloud is directly operated, each tree is identified as a single-tree point cloud entity containing complete space structure and spectrum information, the terrain influence is fundamentally eliminated, the three-dimensional structure parameters of the single tree can be accurately obtained, a two-stage strategy of geometric rough division and spectrum refinement is creatively provided, and the method is suitable for the three-dimensional tree point cloud entity of the single tree. The deep fusion perfectly combines the macroscopic accuracy of geometric segmentation and the microscopic identification of spectrum segmentation, and greatly improves the segmentation precision of the adhered tree crowns.

The invention discloses an energy corridor forest fire risk prediction method and system based on multi-source data fusion, and the method comprises the steps: constructing a fusion data set, and determining risk evolution parameters; if the risk evolution parameter exceeds a preset threshold value, triggering a vegetation difference analysis function to obtain a section heterogeneity label; after section heterogeneity labels are obtained, time-space non-uniformity compensation correction is conducted on historical data through a time sequence change tracking method, the risk transition probability of each section is calculated, and potential critical points are judged; if the potential critical point is judged to be high in probability, activating a difference early warning mechanism, generating a targeted risk level map according to a section heterogeneity label and a risk transition probability, and obtaining a dynamic early warning signal; and carrying out iterative verification on the dynamic early warning signal by adopting a real-time update flow in the fusion data set, adjusting a risk level in combination with prediction output of the long and short-term memory network, and determining a final risk prediction result. According to the invention, the timeliness and accuracy of fire risk prediction are improved.

A long-term continuous learning method based on task core memory management and consolidation aims to enable a model to sequentially learn from a large number of task sequences, new knowledge is obtained, information of previous learning is reserved, and the method is similar to a human learning mode. The method comprises the following steps: 1) task input and instruction fine tuning; 2) performing difference analysis on the model parameters of the current task and the previous task, identifying a task core memory unit, calculating an adaptive weight based on task prototype similarity, and dynamically updating the memory unit; 3) constructing an experience playback buffer area through a difficult sample selection strategy and a difference sample selection strategy; and 4) utilizing the joint loss function training model to keep the memory of the historical tasks while learning the new tasks. According to the method, the problem of disastrous forgetting in long-term continuous learning is mainly solved, and the performance of the model in a long-term sequence task is remarkably improved.