246 results about "Synthetic data sets" patented technology

The invention relates to the field of mechanical equipment health management, in particular to a large sliding bearing fault detection and evaluation method, device and system. Comprising the following steps: collecting multi-source sensing data, and constructing a comprehensive data set; constructing a state space model based on a sliding bearing physical mechanism; the multi-source sensing data and the state space model are fused through Bayesian filtering, and hidden state parameter posterior distribution is dynamically estimated; generating a virtual fault sample by using a generative adversarial network in combination with a physical rule base; designing a Bayesian space-time sequence diagnosis model based on an attention mechanism, and generating fusion health state features; processing and fusing the health state features by using a degradation process model, and predicting the remaining service life of the bearing; and based on the health state, the fault probability and the remaining service life, setting multi-stage early warning threshold values, and triggering intelligent early warning. According to the method, the defect that a single model is insufficient in adaptability and generalization ability under complex working conditions is overcome, and the accuracy and reliability of fault detection are remarkably improved.

The invention discloses an equipment fault intelligent early warning system based on abnormal voiceprint AI analysis of energy equipment, and relates to the technical field of equipment health management, the equipment fault intelligent early warning system comprises an equipment fault early warning platform, and the equipment fault early warning platform is in communication connection with the following modules: a data sensing fusion module, a voiceprint AI analysis module and a voiceprint AI analysis module; the data acquisition module is used for acquiring high-frequency voiceprint signals, temperature field distribution and vibration data in real time during operation of energy equipment through a distributed sensor network to form a comprehensive data set; and the voiceprint AI analysis module is used for extracting voiceprint features from the comprehensive data set, and identifying whether the equipment emits abnormal voiceprints or not by using a pre-trained voiceprint AI model. According to the method, early abnormity is identified through the high-precision AI model, sudden equipment faults are effectively prevented, equipment physical field interaction is simulated in combination with the digital twin technology, and a fault evolution path is dynamically deduced, so that operation and maintenance personnel can take intervention measures at the initial stage of the faults, the stability and reliability of equipment operation are remarkably improved, and the non-planned downtime is shortened.

The invention relates to the technical field of electric power fault diagnosis, and discloses a transformer area line loss abnormity diagnosis method and system based on a large model, and the method comprises the steps: collecting multi-dimensional data used for supporting transformer area line loss abnormity diagnosis, carrying out the cleaning, correlation fusion and standardization processing of the multi-dimensional data, and obtaining a comprehensive data set; a multi-layer diagnosis system based on a rule model, a random forest model and a large model is constructed, and the rule model identifies the simple and conventional anomalies of the transformer area according to a preset anomaly diagnosis rule based on the basic attribute data and the power operation state data in the comprehensive data set; the random forest model locates complex anomalies and novel anomalies by mining a coupling relationship among energy access condition data, external environment influence data and line loss fluctuation in the comprehensive data set; the big model carries out fusion verification on diagnosis results of the rule model and the random forest model, and outputs a final transformer area abnormity diagnosis result; and based on the diagnosis result, a differential loss reduction strategy adaptive to the actual data characteristics of the transformer area is recommended. The working efficiency is improved.

The invention belongs to the technical field of artificial intelligence, and relates to an attribution analysis method and device based on artificial intelligence, computer equipment and a storage medium, and the method comprises the steps: carrying out the information analysis of an alarm notification based on a data perception agent when the alarm notification corresponding to a target system is monitored, and obtaining key entity information; acquiring associated data corresponding to the key entity information based on the data acquisition agent; performing data fusion on the associated data based on the data processing agent to obtain a comprehensive data set; constructing a target prompt text corresponding to the comprehensive data set based on the data interaction agent; based on a large language model, performing association analysis and intelligent reasoning on the comprehensive data set according to the target prompt text, and generating a root cause analysis result; and outputting an attribution analysis result. In addition, the attribution analysis result can be stored in the block chain. The method can be applied to attribution analysis scenes in the financial field and the medical health field, and the processing efficiency and accuracy of attribution analysis are improved.

The invention relates to a grid-connected scheduling management method and device constructed in combination with a knowledge graph, equipment and a medium. According to the method, a comprehensive data set is constructed by integrating multi-source data such as new energy output, power grid topology, load, weather and historical fault records, and then a dynamic knowledge graph is formed by using entity recognition and relation extraction technologies; a probability causal graph model is constructed by extracting a causal path and adding probability parameters, and uncertainty propagation intensity is quantified in combination with a sequence diagram neural network; on the basis of a propagation model, risk index conditional probability is calculated by adopting probability causal reasoning, and a fault propagation sequence is simulated through a cascade failure theory to realize multi-level risk assessment; based on a multi-objective optimization model and deep reinforcement learning, an adaptive scheduling strategy is generated, a complete technical closed loop from data fusion and causal reasoning to intelligent decision is realized, and the technical effects of describing a new energy uncertainty propagation path, prospectively evaluating a power grid risk situation and dynamically generating an optimal grid-connected scheduling scheme are achieved.

The invention relates to the technical field of geophysical exploration, in particular to a multichannel deep learning magnetotelluric inversion method based on physical information constraint. The method comprises the following steps: generating a synthetic data set containing a geoelectric model and forward modeling response thereof, and adding a noise simulation actual observation condition; constructing a hybrid network architecture combining Transform and U-Net, taking apparent resistivity and impedance phase as dual-channel input, extracting global features by using an encoder, gradually recovering spatial resolution through a decoder, and outputting an underground resistivity model; network training adopts a composite loss function fusing model loss and data loss, and an inversion process is constrained by introducing a magnetotelluric forward modeling physical rule, so that a result is ensured to fit observation data and conform to a physical mechanism; after training is completed, preprocessed actual measurement data are input into the model, and a resistivity image can be directly obtained. The method is used for geological structure identification and reservoir interpretation, and the inversion precision and reliability are effectively improved.

Systems and processes are disclosed for enhancing cybersecurity and optimizing software repositories through integration of web crawling, web scraping, feature engineering, and advanced machine learning algorithms to detect phishing attempts, prevent account takeover fraud, and identify unused code in repositories. The system collects and refines data from various sources, including transaction logs, customer databases, device details, external data sources, and historical fraud data, to build comprehensive datasets. Feature engineering creates new, meaningful features from the refined data, which are used to train and evaluate machine learning models. The best-performing models are deployed in production to monitor incoming communications and transactions in real-time, flagging suspicious activities and optimizing codebases. This processing ensures timely detection and prevention of security threats while maintaining efficient software development processes. Robust protection is provided against evolving cyber threats and enhances software performance and security through continuous learning and adaptation.

The invention relates to the technical field of medical evaluation, and discloses an Alzheimer's disease early cognition evaluation system and method based on combination of traditional Chinese medicine and western medicine. The system comprises a tongue image acquisition module, a pulse condition fluctuation signal acquisition module, a western medicine information acquisition module, a data preprocessing module, a feature extraction module, a data fusion module, a comprehensive data set establishment module and an evaluation decision module, wherein the tongue image acquisition module acquires tongue images, pulse condition fluctuation signals, inquiry text data and inquiry sound information; and the western medicine information acquisition module performs multi-modal fusion on feature data of traditional Chinese medicine and western medicine to construct a comprehensive data set. According to the system, multi-modal data is collected by integrating a traditional Chinese medicine four-diagnosis module and a western medicine detection module, the multi-modal data is processed and fused through algorithms including attention mechanism image processing, signal analysis and the like, and then an evaluation model is trained through integrated learning, transfer learning and the like. The method comprises the steps of equipment deployment, data acquisition and processing, model training, evaluation application and the like, multi-dimensional accurate evaluation is realized by combining optimization of a generative adversarial network, fuzzy logic and the like, and technical support is provided for early diagnosis.

The invention discloses an agricultural meteorological disaster monitoring and early warning method and system based on a remote sensing technology, and the method comprises the steps: generating a multi-dimensional data set through obtaining and processing multi-source remote sensing data, and generating a comprehensive data set through weighted average fusion. Then, extracting land surface temperature data, comparing the land surface temperature data with a historical value, calculating a temperature deviation value, and generating temperature anomaly distribution data; and calculating a disaster intensity index by combining the soil humidity and vegetation index data change trend, and generating disaster intensity distribution data. The data meteorology is imported into a driving simulation system, disaster evolution is simulated, and the disaster influence range and duration are predicted. And if the prediction data exceeds an early warning threshold, generating early warning data including disaster categories, influence areas and prediction time, and generating disaster risk distribution data by using a spatial interpolation method for dynamic monitoring. According to the invention, the accuracy and timeliness of disaster monitoring and early warning are improved.

The invention belongs to the technical field of image processing and deep learning, and particularly relates to an image defogging method based on dynamic wavelet prior and double-domain learning. Aiming at the requirements of all-weather clear imaging in the fields of intelligent traffic systems, safety monitoring and the like, and in order to overcome the defect that a static convolution kernel adopted by a traditional defogging method is difficult to adapt to different haze degradation, the invention provides a method for dynamically generating a convolution kernel by using haze priori contained in a multi-scale wavelet LL sub-band; and an efficient, robust and accurate image defogging model is constructed. According to the invention, based on a multi-scale U-shaped coding-decoding architecture, a dynamic wavelet depth separable convolution module DyWConv is embedded in front of each level of a coder to realize content adaptive feature extraction, and a double-domain feature learning module SPAFormer Block cooperatively utilizing Fourier domain global modulation and wavelet domain multi-scale decomposition is designed. And double-domain features are fully fused through an adaptive gating fusion mechanism, and finally a clear image is reconstructed and output step by step. According to the method, a method for explicitly encoding frequency domain degradation prior into dynamic convolution kernel parameters is innovatively provided, the complementary advantages of Fourier transform and wavelet transform are cooperatively utilized, spatial non-uniform haze can be effectively removed, image details can be recovered, leading performance is achieved in a synthetic data set and a real scene, and the method has a wide application prospect.

The invention discloses a differential privacy data set distillation method and system based on image generation data, and belongs to the technical field of data privacy protection and machine learning. The method comprises the following steps: firstly, synthesizing a synthetic data set meeting Gaussian differential privacy, training a feature extractor on the synthetic data set, and finely adjusting an expert model by using original data differential privacy; multiple rounds of iterative optimization are carried out on the distillation data set initialized according to the classes, wherein a feature extractor is randomly selected according to the classes in each round to align the features of the original data added with the noise and the distillation data, and an expert model is utilized to align the semantics of the distillation data hard labels and the semantics of the synthetic data soft labels; and finally outputting a distillation data set meeting the total privacy budget. According to the method, effective priori is provided by utilizing generated data, extra noise injection in the alignment process is reduced, convergence is accelerated, and better privacy protection and data availability balance compared with a previous method is realized under the same differential privacy budget.

The invention discloses a distribution line abnormity monitoring and early warning method and system, and relates to the technical field of distribution line intelligent monitoring. The method comprises the following steps: acquiring operation data acquired by multiple types of sensors, and performing time alignment, normalization and fusion processing to form a comprehensive data set; inputting the comprehensive data set into a generative denoising model based on noise and abnormal signal distribution separability learning, and outputting a clean data sequence through noise suppression and abnormal feature fidelity joint optimization; identifying an abnormal evolution trend with a nonlinear amplification characteristic by using a Lyapunov index and a Hurst index, and generating a risk assessment result; and constructing and dynamically adjusting a self-adaptive early warning threshold set according to a risk assessment result, and outputting abnormal early warning information when a risk index exceeds a threshold. The method realizes parallel noise suppression and abnormal feature fidelity, has dynamic identification and self-learning capabilities, and significantly improves the accuracy of distribution line anomaly detection and the stability of an early warning system.

The invention relates to the technical field of municipal comprehensive GIS data management, in particular to a municipal comprehensive GIS data integration and management system for hub airport construction. The system comprises a data acquisition unit which is used for acquiring spatial geographic parameters, construction parameters and traffic operation parameters to construct an original multi-source data set, preprocessing the original multi-source data set and outputting a standardized data set; the equipment scheduling and resource management unit generates a total conflict set based on the spatial fusion data set and the dynamic progress data set, and obtains a Pareto optimal solution set by constructing a multi-objective optimization function; and the multi-dimensional conflict early warning unit obtains a security conflict set based on the spatial fusion data set and the dynamic progress data set in combination with the total conflict set, and finally generates dynamic early warning information through risk grade division. The coordinated scheduling scheme can be automatically generated under the scene of progress conflict of multiple contractors, complete dependence on manual decision making is avoided, and the construction efficiency and collaboration are improved.

The invention discloses a synthetic data set construction method and electronic equipment, and relates to the technical field of artificial intelligence, and the synthetic data set construction method comprises the following steps: dividing an original multi-source document of a target field into a plurality of word segmentation units by using a word segmentation device; obtaining representative scores of the plurality of word segmentation units on the original multi-source document; based on the representative scores, determining the word segmentation units with the representative scores higher than a first score threshold as candidate keywords; determining importance degree scores of the candidate keywords based on the representative scores of the candidate keywords; based on the importance score, determining the candidate keyword of which the importance score is higher than a second score threshold as a target keyword; and calling a pre-training language model, and based on the target keyword, generating a question and answer pair corresponding to the target keyword to obtain a synthetic data set of the target field. The technical problem that the data coverage rate and the field correlation of the generated synthetic data set are low in the prior art is solved, and the technical effect of improving the data coverage rate and the field correlation of the generated synthetic data set is achieved.

The invention is suitable for the field of education data analysis, and provides a learning cycle data acquisition and evaluation management method and system, and the method comprises the steps: automatically collecting and fusing the full life cycle data of students, and forming a comprehensive data set; dynamically constructing a student digital portrait based on the data set, and constructing a personal knowledge graph taking knowledge points as nodes; in combination with the portrait and the knowledge graph, analyzing relevance between browsing behaviors and course selection and practice, and evaluating a career behavior mode and track; calculating a quantitative career-academic fitness index, identifying an imbalance risk and generating an early warning; personalized resources are pushed to the student terminal based on early warning, early warning is sent to the teacher terminal, and the intervention effect is continuously tracked. The system correspondingly comprises five modules. According to the method, the problems of data islands, portrait statics, lack of fitness and the like in the prior art are solved, and the accuracy, systematicness and effectiveness of learning cycle evaluation and management are improved.

The invention is suitable for the technical field of air conditioners, and provides an air conditioner intelligent regulation and control method which comprises the steps that outdoor environment data, indoor environment data and personnel state data are collected in real time, all the data are fused, and a comprehensive data set is generated; constructing a dynamic temperature gradient field based on the comprehensive data set; dividing functional areas, and calculating an influence relation matrix in combination with the dynamic temperature gradient field; taking the influence relation matrix as a core constraint condition, and adopting a multi-objective optimization algorithm to obtain an air conditioner regulation and control parameter set; and on the basis of the air conditioner regulation and control parameter set, the output parameters of all the air conditioner single bodies are gradually adjusted through a switching mechanism, dynamic global optimization of the air conditioner system on multiple targets such as energy saving, comfort and equipment service life is achieved, and the problems that regulation and control are lagged, energy consumption is too high, the comfort degree is not uniform, and equipment loss is large are effectively solved.

The invention belongs to the technical field of seismic exploration and artificial intelligence, and relates to a physical perception collaborative fusion-based full waveform inversion neural network construction method, which comprises the steps of required model data construction, full waveform inversion neural network construction, loss function construction, network pre-training, and network fine tuning and application. Physical feature perception and collaborative fusion are carried out through a U-Net network and an adaptive residual learning module, so that the problem that traditional full-waveform inversion depends on synthetic data training and is not matched with an actually measured data field is solved. The designed pre-training strategy not only can get rid of dependence on a large-scale synthetic data set, but also can effectively relieve the domain offset problem through fine adjustment of measured data, thereby improving the generalization ability of the model. And the adaptive residual learning module enables the neural network to adaptively and dynamically optimize feature expression, and improves inversion precision through collaborative fusion of a physical perception mechanism. According to the method, the physical consistency and imaging reliability of speed model prediction can be remarkably improved while the labor cost is reduced.

The invention provides a multi-modal data preprocessing and fusion technology and system based on artificial intelligence, and relates to the technical field of data processing, and the technology comprises the steps: converting a PDF document into a table image, and carrying out the preprocessing; a large visual model is adopted for table structure analysis, and zero sample understanding is achieved through a prompt text; finely tuning the model based on the synthetic data set; generating standardized output by using the large language model; and the training lightweight model is deployed on edge equipment to realize offline processing. According to the method, the table identification accuracy is improved, the resource consumption is reduced, and efficient identification and processing of the complex table are realized.

The invention provides a crack identification method and system based on multi-modal data fusion, electronic equipment and a storage medium, and the method comprises the steps: firstly obtaining multi-modal data of a to-be-detected structure surface, specifically including an image sequence obtained by an acquisition module, full-field stress time sequence data and preset key part discrete point stress time sequence data; performing time synchronization calibration on the multi-modal data, mapping a stress field corresponding to full-field stress time sequence data to a pixel coordinate system of an image sequence through camera calibration parameters, and associating discrete point stress time sequence data to the pixel coordinate system to form a comprehensive data set of unified space coordinates; and finally, inputting the comprehensive data set into a trained multi-modal crack recognition model, and performing model recognition to obtain crack parameters and stress analysis results, thereby realizing crack recognition driven by multi-modal data fusion. Multi-modal data collaborative input overcomes the problems of misinformation, missing report and information loss of a single data source, and improves the accuracy and reliability of crack recognition.

The invention provides a method, a device and equipment for identifying vulnerability driving factors of disaster-bearing bodies under typhoon disasters and a storage medium. The method comprises the following steps: obtaining a comprehensive data set according to obtained historical typhoon disaster intensity data and historical typhoon disaster loss characterization data of a plurality of areas; obtaining an optimal function and a target disaster-causing intensity variable according to the comprehensive data set and a plurality of preset candidate functions; according to the optimal function and the target disaster-causing intensity variable, obtaining a disaster-bearing body loss rate and a disaster-bearing body absolute loss value; performing typhoon disaster vulnerability assessment according to the disaster-bearing body loss rate and the disaster-bearing body absolute loss value to obtain a typhoon disaster vulnerability assessment result; and according to the typhoon disaster vulnerability assessment result and a plurality of preset driving factors, determining an identification result of the vulnerability driving factors of the disaster-bearing body under the typhoon disaster. According to the invention, the accuracy of typhoon disaster vulnerability assessment and driving factor identification can be improved, and the pertinence and scientificity of typhoon disaster risk management are effectively improved.

The invention relates to an encoder-RFID positioning inspection robot control method, system and device and a medium. The method comprises the following steps: firstly, acquiring a multi-view image, temperature and environmental gas concentration data of pump room equipment, performing time-space synchronization calibration by combining encoder-RFID positioning data, and fusing into multi-source preliminary fusion data; secondly, robot track data are extracted, the real-time position is calculated, RFID calibration is carried out, and features are extracted and integrated into a comprehensive data set in combination with historical data; analyzing the comprehensive data set to determine a normal mode of the equipment, and comparing parameters to generate an abnormal identifier and a fault early warning signal; and finally, wirelessly transmitting an early warning signal, receiving an instruction, adjusting the state of the robot, returning data and generating monitoring data. By adopting the method, the monitoring and positioning precision can be improved, faults can be early warned, the inspection efficiency is improved, and the operation risk is reduced.

The invention discloses a fire extinguishing method and system of a fire-fighting robot, a medium and a program product of the medium, and relates to the technical field of intelligent fire-fighting robots, and the method comprises the steps that when a fire extinguishing task is executed, a multi-mode sensing module is adopted to recognize a fire source; the multi-mode sensing module captures real-time image data of a fire source area, obtains a thermal radiation intensity distribution diagram of a fire source, and forms a comprehensive data set of the real-time image data and the thermal radiation intensity distribution diagram; the advancing speed of the fire-fighting robot and the comprehensive data set are input into a preset fire extinguishing control model for fire source threat assessment, and fire source threat level feedback information is obtained; performing weight distribution and priority ranking on the comprehensive data set according to the fire source threat level feedback information to obtain an optimization scheme of a fire extinguishing strategy; initial environment data of the target fire extinguishing area is obtained, and an optimal fire extinguishing path and a fire extinguishing agent putting strategy are determined in combination with the optimization scheme and the initial environment data; the environment perception capability and the fire extinguishing efficiency of the fire-fighting robot are improved.

The invention provides an abnormal target detection method and apparatus, an edge computing device and a storage medium. The method comprises the steps of constructing a comprehensive data set based on multi-source image data; training a target detection pre-training model through the comprehensive data set to obtain a target detection training model; and inputting field video monitoring data into the target detection training model, and outputting abnormal information when an abnormal target is detected. According to the technical scheme provided by the embodiment of the invention, the abnormal target is intelligently detected through the edge detection equipment provided with the target detection training model, the real-time monitoring of the working site can be realized without the need of the inspection personnel to arrive at the gas working site for inspection, the real-time performance is high, the labor cost is saved, the accident risk in the inspection work is reduced, and the working efficiency is improved. And the abnormal target detection efficiency is improved.

The invention belongs to the cross technical field of artificial intelligence, satellite remote sensing and insurance science and technology, and particularly relates to a generative AI remote sensing image disaster dynamic monitoring system, equipment and application. Generating a synthetic data set with a physical label by integrating the hydrodynamic model and the building structure dynamic model; a three-source space-time alignment module is used for aligning satellite images, unmanned aerial vehicle video streams and meteorological station data, after three-source asynchronous input data are synchronized, a disaster recognition result of semantic segmentation is output through a wave band self-adaptive attention diffusion model, and then a disaster evolution model of a continuous time domain is constructed through a neural differential equation deduction engine. Outputting disaster diffusion paths and intensity thermodynamic diagrams in future T hours; and automatically generating a disaster damage interval valuation report in combination with OpenStreetMap building semantic information and a regional economic density map. According to the invention, a closed-loop system of physical synthesis, dynamic deduction, insurance pricing and temperature drift correction is constructed, and a monitoring and actuarial integrated solution is formed.

The invention discloses an outdoor space public facility layout optimization method based on big data analysis, and belongs to the field of outdoor space public facility layout optimization, and the method comprises the steps: predicting the number of participants in a data set based on demands, and combining the illumination intensity, wind direction, temperature and rainfall in a comprehensive data set, calculating a meteorological suitability score of each region, and generating a preliminary facility layout data set including candidate position points and meteorological suitability scores thereof; on the basis of the adjusted candidate position data set, in combination with the average decibel value of each region, spatial analysis is executed, position points with the average decibel value lower than 60 decibels are screened, and an optimized position data set is generated and comprises the position points, a meteorological suitability score and a noise level; and based on the optimized position data set, in combination with the facility coverage range and the construction and maintenance cost, calculating the facility number and type of each region, and generating a facility layout planning data set including a facility configuration list of each region.

The invention relates to the technical field of fluid control, and discloses a drip-proof self-adaptive fluid dispensing control method and system. The method comprises the following steps: acquiring an image sequence, pressure data and viscosity change data when fluid is disconnected; positioning an abnormal disconnection moment, extracting a corresponding picture, and extracting a necking diameter and a necking length from the picture; calculating the real-time change rate of the necking diameter based on the necking parameters, obtaining the state identification that the fluid is about to be disconnected, and predicting the disconnection opportunity; integrating the cut-off opportunity predicted value and the viscosity data to obtain a comprehensive data set, and calibrating core parameters to generate an adaptive signal waveform; acquiring a real-time flow velocity data correction waveform, and generating an optimization control signal; and extracting a key intervention point, applying pulse intervention, calculating the colloid residual quantity, performing iterative optimization if the colloid residual quantity exceeds a preset threshold value, and performing no intervention if the colloid residual quantity does not exceed the preset threshold value, thereby finally obtaining a non-leakage control sequence. According to the method, the necking characteristic of the fluid can be accurately captured, the working condition is dynamically adapted, non-dripping dispensing is realized, and the control precision and stability are improved.

The invention provides a data distillation method and device for photovoltaic power generation power prediction and electronic equipment, and the method comprises the steps: obtaining a synthetic data set based on an original data set, and obtaining multi-scale time domain fusion features, frequency domain features and statistical features corresponding to the original data set and the synthetic data set; determining a time domain matching loss function according to the multi-scale time domain fusion features corresponding to the original data set and the synthetic data set; determining a feature domain matching loss function according to the frequency domain features and statistical features corresponding to the original data set and the synthetic data set; combining the time domain matching loss function and the feature domain matching loss function to form a total loss function; and optimizing the synthetic data set by adopting a total loss function to obtain a distillation data set. Based on the method, the distillation data set with rich information can be extracted from the original data set, the data transmission and storage overhead is reduced, and the algorithm development efficiency is improved.

The invention relates to the technical field of construction safety management, in particular to a building construction hazard source dynamic closed-loop management method based on gridding and informatization, which comprises the following steps: performing gridding division on a building construction area through a preset grid division rule, and collecting a building construction hazard source comprehensive data set; preprocessing the building construction hazard source comprehensive data set; inputting the actual hazard source data set into a hazard source identification model for identification; inputting the dangerous source identification result and the actual image block data set into a dangerous source dynamic risk assessment model for analysis; performing information association on the dynamic risk assessment result of the hazard source and the corresponding grid area, and pushing early warning information to a manager terminal of the corresponding grid; and carrying out associated storage on the hazard source disposal record and the building construction hazard source comprehensive data set. According to the invention, multi-modal features can be effectively fused, dynamic risk assessment is realized, and the real-time performance and accuracy of construction safety management are improved.

The invention relates to an urban planning management method, system and device based on big data analysis and a medium, and the method comprises the steps: obtaining original urban planning data of multiple departments, and generating an encrypted data set with spatial attributes through homomorphic encryption and spatial grid coding; injecting antagonism abnormal data in the ciphertext space to generate a spatial correlation synthetic data set; normal and abnormal event prediction gradients are fused based on double-branch federal training, and a global prediction model is constructed; the privacy risk and the model precision are dynamically balanced through reinforcement learning, and the fusion weight and the synthesis proportion parameter are optimized; and solving the optimal solution of the infrastructure layout in combination with spatial constraints, and generating an encryption planning scheme. The method breaks through the limitation that privacy protection and data utility are difficult to consider in the traditional technology, solves the problems of abnormal simulation failure caused by encrypted data distortion, precision-safety contradiction caused by static parameter stiffness and lack of multi-target collaborative optimization of a planning scheme, and improves the scientificity and adaptability of urban planning decision.

The invention discloses a new energy fan maintenance operation safety monitoring method combined with big data analysis, and relates to the technical field of big data analysis, and the method comprises the steps: collecting new energy maintenance data, carrying out the time synchronization, format standardization and space alignment processing, and forming a unified comprehensive data set; inputting the comprehensive data set into a pre-trained high-fidelity digital twinborn model, mapping risk factors in the comprehensive data set into a three-dimensional space of the digital twinborn model through a built-in risk quantification algorithm, and outputting a dynamic risk field; and associating the risk points and the intensity values in the dynamic risk field as nodes and attributes to nodes and attributes corresponding to a predefined causal knowledge graph, deducing a potential risk evolution path and occurrence probability, and generating a risk deduction conclusion report. According to the method, path exploration and stochastic simulation are carried out through the causal knowledge graph in combination with Monte Carlo tree search, and probabilistic deduction of the potential risk evolution path in the fan maintenance operation is realized.