416 results about "Cloud modeling" patented technology

The invention relates to the technical field of BIM, in particular to a BIM-based construction progress intelligent management and control method and system, and the method comprises the steps: constructing a BIM digital twin cloud model, fusing a panoramic image and LiDAR point cloud data, dynamically selecting a data source update model through an algorithm, extracting engineering topology, building a construction network diagram, and generating a state evolution trajectory in combination with environment and resource data. Predicting and visualizing milestone time, comparing field data with a BIM model to generate progress deviation information, performing stability analysis, triggering resource allocation when a threshold value is exceeded, synchronizing a material supplier and a field manager, dynamically adjusting personnel, equipment and materials, dynamically adjusting a milestone plan based on multiple data, and presenting and guiding allocation through the BIM model. The construction progress is ensured to be consistent with the plan, and the complex interaction relationship and dynamic characteristics in the construction process are captured through nonlinear dynamic system modeling in combination with LiDAR point cloud and other high-precision data.

The invention discloses an industrial inspection intelligent decision-making method and system based on large and small model collaboration. The industrial inspection intelligent decision-making method comprises the following steps: dividing an inspection task into a plurality of sub-tasks by using a cloud large model, and performing dynamic task planning based on the sub-tasks; calling an edge small model to execute the subtask to obtain multi-modal original data; performing cross-modal feature fusion on the multi-modal original data by using a cloud large model to obtain cross-modal fusion features; and reasoning based on the cross-modal fusion features by using a cloud large model, extracting abnormal factors, and formulating a new inspection task based on analysis of the abnormal factors. Through the comprehensive method of comprehensive multi-mode perception, conditional diffusion model enhancement and cloud-edge collaborative decision, the problems of fault sample scarcity and perception deviation in an extreme environment are effectively solved, high precision and low time delay are considered, and a new breakthrough of intelligent inspection is brought.

The invention discloses a dynamic data closed-loop system based on real-time positioning confidence evaluation and an optimization method. The dynamic data closed-loop system and the optimization method are suitable for high-precision positioning scenes such as automatic driving and unmanned systems. The system collects data of a GPS, an IMU, a laser radar and a camera through a multi-source sensor, and a confidence evaluation module is adopted to fuse sensor data quality, environment interference indexes and historical track consistency to calculate positioning confidence. Based on confidence score, the system triggers different data acquisition strategies and sensor parameter configurations, and dynamically optimizes resource allocation. The system comprises an edge end fusion optimization module and a cloud model iteration module, constructs a closed loop mechanism, uploads data to a cloud for model updating and returns parameters when the confidence coefficient is abnormal, and improves the adaptability and robustness of the system. Experiments show that the method can effectively reduce the positioning error, optimizes the use of computing resources, and has good real-time performance, expansibility and engineering feasibility.

The invention relates to the field of data management, in particular to a VOCs pollution working condition data treatment method based on artificial intelligence. The method comprises the following steps: firstly, constructing a three-dimensional model of a factory, a pipeline and treatment equipment based on a BIM tool and a GIS technology, and deploying an Internet of Things sensor to collect VOCs pollution data; collecting original infrared absorption spectrum data of VOCs, and performing component identification by using a non-negative matrix factorization algorithm; a machine learning model is trained in combination with historical data, and short-term emission prediction is carried out; cooperative scheduling is carried out on the governance equipment through a multi-agent game optimization algorithm, and an optimal operation scheme is generated; constructing volume cloud modeling, and dynamically displaying an optimization effect; and if the predicted emission exceeds the standard, performing emission abnormity traceability analysis by using a graph neural network, and outputting a fault diagnosis report. According to the method, the VOCs emission prediction precision and the treatment efficiency are effectively improved.

The invention relates to the technical field of intelligent assessment, and discloses a dam risk assessment method, device and equipment based on multi-source data fusion, and the method comprises the steps: determining a monitoring project type of dam multi-source monitoring data, building a risk assessment index system through selecting a key index of the monitoring project type, and obtaining a risk assessment result; the method comprises the steps of dividing a dam region by combining measuring point spatial distribution to obtain a dam partitioning result, processing single measuring point actual measurement data of the same region in the dam partitioning result through a dam risk assessment model to obtain a region basic probability distribution value, and further obtaining a risk assessment result. According to the method, the limitation of single data dimension is broken through by constructing a risk assessment index system, accurate mastering of risk heterogeneity of different regions is realized through region division, a dam risk assessment model combining a cloud model and an evidence theory is adopted, the problem of dimension unification of multi-source data is solved, the risk assessment comprehensiveness is ensured, and the risk assessment efficiency is improved. And the accuracy of risk assessment is also improved.

The invention discloses a flexible automatic production management and control system for a new energy high-power high-frequency transformer, and relates to the field of production management and control, the flexible automatic production management and control system comprises a production demand receiving module, a production resource management module, a production planning module, a production execution module and a quality control module, an order is received through an API or EDI, and a work order is generated; collecting equipment and material information in real time, and constructing a production resource association network; carrying out production scheduling and process matching by adopting an improved deep learning model; scheduling materials, procedures and equipment based on a multi-level analysis control model; the quality is evaluated by using a two-dimensional cloud model, and data recording is performed through a distributed database, so that the traceability of the quality problem is ensured. According to the system, the production efficiency and the quality management level are effectively improved, the problem that existing rigid automatic equipment lacks flexibility is solved, and efficient and accurate production of the new energy high-power high-frequency transformer is ensured through cooperative work of a plurality of modules so as to meet different production requirements.

The invention relates to the technical field of rail traffic equipment maintenance and artificial intelligence image recognition, in particular to a rail vehicle cloud edge cooperative detection method and system and inspection equipment. The method comprises the following steps: acquiring image data of an area to be detected, synchronously recording attitude information of shooting equipment, carrying out image quality scoring on the image data, if the image quality score reaches a preset threshold value, calling a local target detection model to carry out defect identification on an image, and outputting an identification result containing a target type, a position and confidence; if the confidence coefficient of the recognition result is lower than a set threshold value, the image data and the posture information are uploaded to a cloud server, a cloud high-precision recognition model is called for secondary recognition, and a cloud recognition result is obtained. The real-time requirement is met through quick response of the local model, the cloud model rechecks a low-confidence result, efficiency and precision are both considered, and the contradiction that in the prior art, manual judgment standards are different, and speed and precision are difficult to consider is solved.

The invention provides an equipment abnormal voiceprint detection system. The equipment abnormal voiceprint detection system comprises a voiceprint collection module used for collecting original voiceprint signals in real time and preprocessing the original voiceprint signals; the edge end detection module is deployed on edge computing equipment and is used for carrying out real-time anomaly detection on the voiceprint features by utilizing a one-dimensional lightweight neural network model; the abnormity credibility evaluation module is used for converting a real-time abnormity detection result into a probabilistic abnormity credibility score; the incremental data screening module is used for screening high-value samples from the real-time voiceprint data based on a dynamic density trend sensing algorithm and caching the high-value samples in the edge; the cloud model evolution module is used for carrying out evolution training on the detection model by utilizing a continuous learning mechanism; the generation and playback module is used for jointly generating pseudo samples through a variational auto-encoder VAE and a generative adversarial network GAN; and the model updating module is used for compressing the evolved cloud model and then issuing and replacing the original model in the edge end detection module so as to form a cloud-edge collaborative sustainable evolution closed loop.

The invention relates to the technical field of intelligent monitoring of petroleum equipment, and particularly provides an industrial equipment state diagnosis method based on multi-modal fusion deep learning. The method comprises the following steps: synchronously acquiring load, vibration and current data through an explosion-proof RTU (Remote Terminal Unit), and transmitting the data to an edge node through an industrial LoRaWAN (Load RaWide Area Network); aligning a multi-source signal time sequence by adopting Kalman filtering, and fusing to generate data streams with consistent time domains; features are extracted by using wavelet packet energy entropy, and a lightweight ResNet18 model is input to realize high-precision diagnosis; based on the diagnosis result, a dynamic optimization strategy is generated through an LSTM-Transform parallel model; and finally, issuing a control instruction to the equipment for execution, and collecting feedback data to update the cloud model. Through the multi-rate signal time sequence alignment and feature-decision two-stage fusion technology, the problem of diagnosis errors caused by data asynchronization in a traditional method is solved, the fault recognition precision and the equipment energy efficiency are remarkably improved, and meanwhile the requirements for explosion prevention and real-time performance of the industrial environment are met.

The invention discloses a sensitive cue word generation method and system based on parameter sensitivity quantification, relates to the technical field of artificial intelligence security, and aims to solve the problem that sensitive test samples are difficult to generate in integrity verification of a black box large language model. The method comprises the steps of obtaining an original large language model; a comprehensive parameter sensitivity index is constructed, the microscopic sensitivity used for representing tiny parameter modification and the macroscopic sensitivity used for representing large modification such as parameter quantification and pruning are fused, and therefore the sensitivity of cue words is comprehensively quantified; in the continuous embedding space, a gradient optimization algorithm is adopted to maximize the comprehensive index as a target to carry out iterative optimization, and constraints such as semantic rationality are applied to ensure that the generated cue word is natural and smooth; and finally, mapping the optimized embedded vector back to the discrete lexical element sequence to obtain a final sensitive cue word. According to the method, the low-cost and automatic test sample generation is realized, and the accuracy, efficiency and concealment of cloud model integrity verification are remarkably improved.

The invention relates to an end-cloud collaborative video abnormity alarm method and system, and the method comprises the steps: obtaining a video stream; generating end-side detection metadata by a first model of the end-side device; when a preset rule is triggered, executing intelligent event encapsulation to generate event data; the event data is uploaded to a corresponding cloud server through a side device according to the risk level through an elastic pipeline; generating text description information matched with the end-side metadata through a second model of the cloud server; wherein the first model and the second model are generated based on joint adversarial training, and semantic consistency in a severe environment is ensured; generating multi-modal information according to the end side detection metadata and the text description information; and performing key target processing on the multi-modal information to output alarm information. Through the method and the device, the problem of description confusion caused by end cloud model judgment contradiction and cooperative splitting in traditional video monitoring is solved, the anomaly recognition accuracy is improved, and security decision is promoted to be transformed to precision and intelligence.

The invention discloses an intelligent risk early warning method and system based on multi-source data fusion, and belongs to the technical field of construction risk assessment and early warning, and the method comprises the steps: obtaining multi-source information, and obtaining an early warning control quantity based on the multi-source information and a preset quantification processing mode; constructing an initial early warning cloud picture based on the early warning control quantity and a preset grading early warning standard; obtaining a three-dimensional early warning cloud picture based on the initial early warning cloud picture, the two-dimensional matrix distribution and a preset two-dimensional normal cloud model; optimizing and correcting the three-dimensional early warning cloud atlas by adopting a posterior probability support vector method to obtain an improved prediction model; and obtaining a risk early warning result based on the improved evidence fusion method and the improved prediction model. According to the intelligent risk early warning method and system based on multi-source data fusion provided by the invention, a corrected three-dimensional early warning model is constructed by combining historical data on the basis of multi-source data fusion and adopting a posterior probability support vector method and an improved evidence fusion method, and the accuracy, effectiveness and applicability of risk early warning are remarkably improved.

The invention provides a cloud intelligent gas leakage analysis and decision-making method based on lightweight discrimination, and the method comprises the steps: enabling a mobile inspection platform to movably collect the gas leakage monitoring information of each monitoring point in a monitoring region, and determining a lightweight discrimination index; the mobile inspection platform determines a hard threshold discrimination result by adopting a hard threshold decision rule based on the lightweight discrimination index; if it is judged that the risk level is high, local response is immediately carried out, and a preset decision scheme is matched; if the risk level is judged to be low, edge node region decision is started; an edge node calculates a gas concentration accumulated value, a concentration fluctuation intensity degree and a sudden increase leakage amplification early warning factor, the gas concentration accumulated value, the concentration fluctuation intensity degree and the sudden increase leakage amplification early warning factor are input into a LightGBM time sequence classification model together with a lightweight discrimination index, a leakage risk level is determined according to model output and serves as a soft threshold discrimination result, and a preset decision scheme is matched; and the cloud model optimizes and collects conflict samples to carry out incremental training on the model of the edge node. According to the invention, the missed alarm rate can be reduced, and the discrimination robustness is improved.

The invention discloses a pump station unit state comprehensive evaluation method, and belongs to the technical field of pump station unit detection. The method comprises the following steps: acquiring multi-source monitoring data and generating a standardized monitoring data set; time-varying mutual information between indexes is calculated, a dynamic threshold value is determined in combination with current working condition parameters and a water level difference correction term, and a dynamic association network is constructed; performing physical mechanism characteristic decoupling on the data, including stripping a vibration signal working condition drift component based on a reference curve to obtain a vibration residual error, and calculating an equivalent standard working condition temperature based on a heat balance principle; respectively calculating a first weight based on data statistics and a second weight based on network topology, and adaptively generating a comprehensive coupling weight according to a consistency coefficient of the first weight and the second weight; and finally, judging a health state level by using a cloud model. Through deep fusion of a physical mechanism and data driving, the problems that fault features are difficult to extract and the model robustness is poor under variable working conditions are solved, and accurate evaluation of the unit state is achieved.

The invention relates to the technical field of medical health monitoring, in particular to a portable intelligent breathing mask and a remote monitoring method thereof. The method comprises the steps that airflow, humidity, blood oxygen and temperature data in the mask wearing process of a user are obtained, and time sequence alignment and feature fusion are carried out; inputting the fused physiological feature sequence into an embedded neural network model to recognize respiratory behaviors, and generating an abnormal probability value; carrying out compression coding based on the parameters, setting an uploading period according to a self-adaptive formula, and uploading the data to a cloud end; the cloud model further analyzes and generates health assessment parameters and a risk trend sequence; and dynamically adjusting an early warning threshold according to an evaluation result, and updating a local model and a sampling strategy in combination with user feedback. The portable breathing mask has the functions of fusing multi-modal data analysis, edge-cloud collaborative recognition, self-adaptive uploading scheduling, personalized early warning regulation and control and the like, and the intelligent level and practicability of the portable breathing mask in remote health monitoring are improved.

The invention discloses a combined electric appliance health analysis method and system based on multi-source data fusion, and the method comprises the steps: cleaning text data, carrying out the semantic mapping of the text data to a preset state quantity index, extracting a basic feature value from image data, and generating a standardized evaluation index set; calculating a weight vector of each evaluation index by constructing a judgment matrix, and performing consistency verification; weibull distribution fitting is carried out based on historical state quantity data, and grading threshold values of the positive degradation index and the negative degradation index are calculated; dividing a state interval based on the normal cloud model and calculating a membership degree vector of each state level; the membership degree vector is converted into a basic probability distribution function, evidence fusion is carried out in combination with the weight vector, and a comprehensive state level is output; and identifying the high-conflict evidence according to the Pignistic probability distance, performing refusion after correction, and outputting a final state level. Deep fusion and intelligent evaluation of multi-source heterogeneous data are realized, and the accuracy and robustness of health state judgment of the combined electric appliance are improved.

The invention relates to the field of laser cutting control, in particular to an autonomous controllable high-power laser two-dimensional cutting control method and system. The method comprises the following steps: acquiring a real-time monitoring image of a to-be-cut steel structure workpiece and a preset cutting control log; performing three-dimensional form point cloud modeling on the real-time monitoring image of the to-be-cut steel structure workpiece to construct a three-dimensional workpiece form model; full-stage cutting demand mining is carried out based on a preset cutting control log, and self-adaptive laser power adjustment is carried out, so that self-adaptive laser power parameters are obtained; carrying out multi-time-point cutting form demand analysis and dynamic cutting path planning on the three-dimensional workpiece form model so as to generate a multi-time-point dynamic cutting path; and real-time cutting control is conducted according to the self-adaptive laser power parameters and the multi-time-point dynamic cutting path, cutting track dynamic optical flow tracking is conducted, and a time sequence laser cutting track sequence is constructed. According to the laser two-dimensional cutting control system, efficient and accurate laser two-dimensional cutting control is achieved.

The invention relates to a roof complex curved surface curvature analysis and optimization method and device, equipment and a medium, and the analysis and optimization method comprises the steps: obtaining original point cloud data of a target roof, carrying out the preprocessing of the original point cloud data, and generating a corresponding roof entity point cloud model; establishing a curved surface reconstruction algorithm interface of the roof entity point cloud model, extracting a corresponding space coordinate point set, and generating a corresponding curvature parameter set; inputting the curvature parameter set into a preset elastic mechanical model, and determining a corresponding critical region; performing parameterization adjustment on the critical region to generate optimized curvature data, and converting the optimized curvature data into plate rolling process parameters to generate a corresponding optimization scheme; and pushing the optimization scheme to each application terminal based on the cloud collaboration platform. The precision and efficiency of modeling and analysis of the complex curved surface are improved, closed-loop cooperation of design, analysis and manufacturing is achieved, and the problems of modeling distortion, mechanical analysis lag and data isolation in a traditional method are solved.

The invention discloses an environment monitoring method based on a low-altitude aircraft, belongs to the technical field of low-altitude aircrafts, and realizes high-precision monitoring through cooperation of anti-interference hardware, edge calculation, navigation communication and cloud modeling. According to the system, a hexagonal prism cabin body is integrated with multiple types of sensors, and data are dynamically calibrated through a BP neural network. And aligning the multi-source spatio-temporal data by using extended Kalman filtering, and generating a three-dimensional grid model containing obstacles and a flow field. Navigation adopts Beidou / IMU combination and visual SLAM redundancy switching, and communication is guaranteed through a 5G private network and UWB double links. And the cloud constructs a three-dimensional AQI thermodynamic diagram based on inverse distance weighting and a Kriging algorithm, predicts pollutant diffusion in combination with a random forest, and drives the aircraft to execute a spiral / grid composite route. The problems that traditional monitoring is low in density, complex in model and insufficient in precision are solved, and a full-link anti-interference perception-decision closed loop is achieved.

The invention discloses a head reconstruction method based on Gaussian sputtering, and belongs to the field of computer vision. Performing alignment and clone splitting on the initial point cloud model by using the supervision image to obtain a head accurate point cloud of the supervision image, and taking the head accurate point cloud position as a Gaussian point cloud position; projecting the Gaussian point cloud to an image coordinate system, and performing feature extraction by taking the supervised image as convolutional neural network input to obtain a convolutional feature map; sampling the head precise point cloud and the convolution feature map of the image coordinate system to obtain a feature vector, and decoding the feature vector to obtain a Gaussian point cloud parameter so as to obtain a three-dimensional head Gaussian model; a head image of any viewpoint is quickly rendered through the three-dimensional head Gaussian model, pixel-by-pixel loss calculation is performed on the rendered image and a real image, and the three-dimensional head Gaussian model is optimized. According to the method, under the condition of sparse input, the convolutional neural network and Gaussian sputtering are fitted, and the obtained high-fidelity three-dimensional head Gaussian model can be rendered quickly in real time and has generalization.

The invention discloses a photovoltaic module health state real-time monitoring method, system and device and a medium, and relates to the technical field of photovoltaic energy monitoring and intelligent operation and maintenance, and the method comprises the steps: collecting the health data of a photovoltaic module through a photovoltaic module working state sensor and an environment sensor, and transmitting the health data through a multi-network protocol; a feature importance evaluation method is adopted to construct a priority scheduling model to dynamically adjust the photovoltaic module health data acquisition frequency, anomaly detection is performed on the acquired photovoltaic module health data on an edge device, a graph convolutional network and a time sequence prediction model are fused to construct a cloud model, an anomaly detection result is input into the cloud model, and the cloud model is subjected to feature importance evaluation. Health trend analysis and state prediction are executed through the cloud model; early warning grading processing is carried out according to health trend analysis and state prediction results, a multi-level response mechanism is established through a correlation analysis algorithm, and a maintenance resource linkage process is triggered. According to the method, the real-time performance of key data acquisition and the data transmission efficiency are effectively improved through a priority scheduling scheme based on the information entropy weight.

The invention provides a winter wheat soil moisture inversion method and system based on an improved water cloud model, storage equipment and electronic equipment, and the method comprises the following steps: obtaining total backscattering characteristics VV and VH, an incident angle and a plurality of polarization characteristic parameters through employing a single-view complex image and a geocoding radiation correction image; calculating a vegetation index from the preprocessed optical image by applying a wave band calculation mode; an improved water cloud model method is applied to correct the radar backscattering coefficient affected by the winter wheat plants, the backscattering coefficient of the earth surface is obtained, and data redundancy is reduced; performing hyper-parameter optimization of the CNN network by applying a CPO algorithm; and constructing a CNN-CPO-RF model by using CNN and RF models to carry out soil moisture estimation. According to the invention, a more accurate soil moisture result can be obtained. According to the method, the robustness of farmland surface soil moisture inversion is remarkably improved, and a new technical means is provided for performing soil moisture inversion by using a satellite-borne SAR image.

The invention discloses a multi-mode continuous transportation supply and demand matching degree evaluation method for a comprehensive transportation junction, and the method comprises the steps: carrying out the collection, cleaning, standardization and time-space alignment of multi-source passenger flow and transport capacity data of inter-city traffic and an intra-city continuous mode, and constructing a unified data basis; according to operation characteristics of different connection modes, a differential supply and demand measurement model is established, and accurate mapping of multi-mode passenger flow and transport capacity is realized through a total passenger flow-connection mode distribution matrix; and introducing a cloud model theory, and carrying out weighted fusion on matching degree indexes of each continuing mode to describe fuzziness and uncertainty in a supply-demand relationship, so as to realize quantitative evaluation of the multi-mode transport capacity coordination level of the comprehensive hub. Through differential supply and demand modeling and cloud model fusion, the supply and demand matching state of multi-mode continuous transportation of the comprehensive passenger transport hub can be accurately reflected, and a scientific decision basis is provided for operation scheduling and resource optimization of a hub management department.

The invention discloses a civil engineering foundation pit deformation monitoring system based on image analysis, and relates to the technical field of foundation pit deformation monitoring, and the system collects and enhances a multi-modal image, eliminates interference, corrects image distortion, tracks key points, constructs a panoramic image, generates a point cloud, extracts deformation features, and constructs a deformation map. The method comprises the following steps: identifying a construction disturbance event, constructing a deformation chain and performing causal reasoning, identifying a high-risk area, predicting a deformation trend and outputting an early warning, displaying a monitoring result, collecting user feedback and synchronizing system data. A deformation track is accurately extracted through image correction and key point tracking, a three-dimensional deformation map is constructed in combination with point cloud modeling, causal analysis is carried out based on construction disturbance and a deformation path, a high-risk area is predicted and identified by using a time sequence trend, an early warning is given out, user feedback collection and result synchronization are supported, and monitoring intelligence and response efficiency are improved.

The invention belongs to the technical field related to tunnels, and provides a geological risk-torque prediction driven jamming risk evaluation method and system in order to solve the problems that risk prediction is inaccurate and early warning lags and the like due to the fact that an existing monitoring method is difficult to achieve collaborative perception of geological inducements and equipment responses in the tunneling process, and the geological risk-torque prediction driven jamming risk evaluation method and system are provided. The method is realized by constructing a standard cloud model, and a dynamic adjustment mechanism based on real-time data fluctuation and risk sensitivity is introduced, so that the cloud model not only can more accurately respond to complex and dynamically changing geological conditions, but also can keep efficient and accurate risk assessment capability during geological sudden change; by adopting a geological risk prediction and torque prediction dual-drive prediction method, collaborative perception of geological inducements and equipment response is realized, the cause and evolution process of the jamming risk are comprehensively and dynamically mastered, and the prediction accuracy is ensured.

The invention discloses a cross-basin water transfer project scheduling scheme evaluation method and system based on a cloud model, and the method comprises the steps: data preparation: collecting scheduling scheme data, and carrying out the normalization processing, and obtaining a reference database; standard cloud construction: defining a comment set, calculating standard cloud digital features and generating a standard cloud; comprehensive cloud construction: evaluating the scheduling scheme by using an expert scoring method, obtaining expert scoring data, calculating through a reverse cloud generator to obtain an evaluation cloud, determining a weight coefficient of an evaluation index by using an entropy weight method in combination with the reference database, and generating a comprehensive cloud; and scheme evaluation: establishing quantitative association of the standard cloud and the comprehensive cloud through dynamic scheme mapping, and determining a scheduling scheme evaluation grade through the similarity of the comprehensive cloud and the standard cloud. The method effectively processes the interactive influence of hydrological randomness and engineering regulation fuzziness, realizes qualitative and quantitative conversion, and improves the scheduling efficiency. The scheduling scheme can be evaluated objectively and scientifically, and a reliable basis is provided for a cross-basin water transfer decision.

The invention discloses an Internet of Vehicles low-delay federated learning method based on semi-asynchronous communication, and the method comprises the steps: constructing a vehicle-road cloud cooperative federated learning framework and a system performance model, building a federated learning training time delay optimization model, minimizing the overall training time delay, and meeting the vehicle energy consumption upper limit and global model precision constraint conditions; a multi-dimensional priority dynamic vehicle selection mechanism is realized, and probability sampling is carried out by using a roulette selection algorithm; a vehicle-RSU federated learning strategy based on semi-asynchronous communication and knowledge distillation optimizes the problem of lagging behind and the problem of model obsolessness; a cloud model aggregation opportunity is optimized based on a model difference metric and an adaptive cloud aggregation strategy of a dynamic threshold trigger mechanism. According to the method, collaborative optimization is carried out on multiple levels of equipment selection, intra-layer communication aggregation, old model processing, inter-layer aggregation triggering and the like, so that the overall training time delay of the Internet of Vehicles federated learning system can be effectively reduced on the premise of satisfying model precision and equipment energy consumption constraints.

The embodiment of the invention provides an insurance loss assessment method and device, electronic equipment and a medium, belongs to the technical field of image modeling, and is applied to financial scenes. The method comprises the steps of performing change area detection according to a current three-dimensional point cloud data set and a historical three-dimensional point cloud data set, constructing a change point cloud model of a change area point cloud data set, and performing point cloud registration on point cloud in the change point cloud model and the historical three-dimensional point cloud data set, carrying out differential processing on the registered point cloud data set and the historical three-dimensional point cloud data set to obtain target change point cloud data; and performing insurance loss assessment on the target loss assessment object according to the target change point cloud data. According to the embodiment of the invention, the target change point cloud is determined through the change area detection, the change point cloud model, the point cloud registration and the point cloud differential operation of the target loss assessment object, and the insurance loss assessment is carried out according to the target change point cloud data, so that automatic loss assessment can be realized, manual intervention is not needed, and the efficiency of insurance loss assessment is improved.

The invention discloses a collaborative safety evaluation method and system for a water conservancy project dam and a slope, and relates to the technical field of water conservancy project safety monitoring and evaluation, and the method comprises the steps: constructing a safety evaluation index system based on a structure overall deformation index, a seepage index, a joint crack index and a material response index; based on a safety evaluation index system, a multi-source monitoring data set is acquired, reverse modeling is performed by using a cloud model, a t-Copula function is introduced to describe a tail joint dependence structure of each index under an extreme working condition, and the overall operation state of the structure is judged by calculating a multivariable collaborative membership degree and combining a standard grade cloud picture. According to the method, the limitation on index independence hypothesis in a traditional evaluation method is broken through, the capability of identifying the collaborative instability mode under the extreme working condition is achieved, the robustness of safety evaluation and the sensitivity of early warning are remarkably improved, and the method is suitable for risk identification and intelligent diagnosis of water conservancy project dams and slopes.

The invention discloses a heterogeneous computing power resource allocation method and system, and belongs to the technical field of computing power scheduling. The method comprises the following steps: analyzing an edge task through a content value analysis model, and generating a task label containing real-time, accuracy and exploratory demand scores; based on the task labels, a distributed computing power distribution protocol is utilized to match and distribute computing power resources for the tasks, and primary distribution and execution are completed; in the task execution process, the data value is evaluated through the lightweight evaluation model; if the evaluation value exceeds a dynamic threshold value, a computing power scheduling model agent is triggered to perform centralized secondary distribution, and the agent decides an optimal uploading path based on deep reinforcement learning and adjusts the threshold value so as to efficiently upload high-value data to a headquarter cloud and drive a large model to autonomously evolve. Through the double-track parallel architecture, the resource conflict between the real-time guarantee of the edge task and the evolution of the cloud model is solved, and the efficient and self-adaptive allocation of the computing power resource is realized.