2623 results about "Iterative refinement" patented technology

The invention relates to the technical field of smart power grids and artificial intelligence, in particular to a substation autonomous intelligent inspection method and system based on multi-modal data, and the method comprises the steps: obtaining inspection data of multiple modals, and generating fusion features; identifying system alarm information; performing intention recognition and task classification to generate an executable task sequence; generating a multi-device cooperative scheduling scheme; executing the multi-device cooperative scheduling scheme; iterative optimization is carried out; according to the intelligent inspection method provided by the invention, more accurate and more robust multi-mode perception and diagnosis are realized, and deep understanding of complex instructions and safe and efficient cooperation of multiple devices are also realized; and meanwhile, through dynamic re-planning and a verification type feedback learning mechanism, high real-time performance and robustness are ensured, and meanwhile, the system is endowed with the capability of iterative optimization.

The embodiment of the invention provides a computer communication method and system based on the Internet of Things, and the method comprises the steps: constructing a multi-level system architecture, carrying out the preprocessing and feature extraction of an original data flow, recognizing the data characteristics through a time sequence analysis method, and constructing a dynamic data model; deploying a monitoring agent at a transmission node to collect network performance indexes in real time, and constructing a network quality evaluation model; for an incomplete sensor data flow, prior probability distribution is constructed based on a dynamic data model and a network quality grade, and an optimal estimation value of missing data is calculated by adopting a Bayesian reasoning framework and an iterative algorithm; establishing a mapping relation between a network state and an optimal parameter through reinforcement learning to realize self-adaptive adjustment and optimization; and grading the data according to reliability, extracting high-reliability data points as anchor points, designing an iterative refinement algorithm to realize information propagation, and fusing to obtain a complete sensor data stream. According to the method, the problems of poor data recovery accuracy, static parameter configuration and insufficient adaptability in a complex network environment are solved.

The invention discloses a carbon emission real-time regulation and control method and system based on multi-source data fusion and AI decision, and relates to the field of data processing systems or methods specially suitable for administrative, commercial, financial, management, supervision or prediction purposes, in the method, real-time production data is collected through Internet of Things equipment, and the real-time carbon emission intensity is calculated in combination with a carbon accounting engine. And constructing an AI optimization layer comprising a process knowledge graph, a long-short-term memory network and a reinforcement learning agent, converting process constraints into mathematical boundary conditions, predicting energy demands, carrying out iterative optimization by taking carbon emission intensity minimization as a target, generating a dynamic scheduling strategy, and realizing real-time adjustment of operation parameters of production equipment. The method and the device are used for improving the accuracy of carbon emission monitoring data on production scheduling and reducing the risk of high-carbon-intensity production caused by improper time period selection of enterprises.

The invention relates to the technical field of neural network model construction, in particular to a lightweight neural network model construction method, which comprises the following steps: based on a target task data set, in a lightweight basic operator library comprising a depth separable convolution, an inverted residual structure and an attention mechanism module, constructing a lightweight neural network model; and searching and jointly optimizing network structure parameters and weight parameters through the differentiable neural architecture to obtain an initial lightweight network model. And deploying the initial model in a target hardware simulation environment, and generating a Pareto optimal model cluster through structural re-parameterization and hardware-aware progressive channel pruning iterative optimization by taking model precision, reasoning delay and memory occupancy as collaborative optimization targets. And selecting a reference student model from the clusters according to deployment constraints, constructing a distillation framework taking the initial model as a teacher model, and performing fine adjustment by adopting a mixed strategy fusing multi-dimensional distillation loss to obtain a final model. The model gives consideration to precision and efficiency, and the detection efficiency and the quality control level are improved.

The invention provides a code generation and evaluation method and system based on RAG and a multilevel decision tree, and the method comprises the steps: integrating project related design documents, and constructing a knowledge base capable of semantic retrieval through a vectorization technology; associating business demand description with related documents in the knowledge base based on an RAG technology, and performing demand semantic enhancement to generate a technology demand cue word; receiving the technical requirement cue word by adopting a large language model so as to generate a complete code conforming to business logic; constructing a four-level decision tree evaluation system, and sequentially executing code quality scanning, deployability verification, dynamic test verification and demand satisfaction verification through an evaluation assembly line to generate an evaluation result; and generating an optimization suggestion according to the evaluation result so as to trigger an iteration generation process when the code does not pass the verification, thereby solving the problems of disjunction between code generation and business requirements, low verification efficiency and insufficient iteration optimization.

The invention discloses a surface mine digital safety production management method and device and a medium, and relates to the technical field of intelligent mines, and the method comprises the steps: collecting rock mass physical and mechanical parameters of a gold mine mining area and three-dimensional geometric morphology data of a blasting working face, and taking the parameters as physical constraint conditions to construct a digital twinborn body; a preset target blasting scheme is called to be input into the element reinforcement learning agent, multi-round simulation deduction is carried out through a blasting scheme evaluation strategy, and a simulation deduction result is generated; when the peak particle vibration velocity in the simulation deduction result exceeds a preset peak particle vibration velocity threshold value, generating a parameter adjustment suggestion; and the parameter adjustment suggestion is fed back to the element reinforcement learning agent for iterative optimization until the peak particle vibration velocity does not exceed a preset peak particle vibration velocity threshold, and a blasting scheme is generated. Through autonomous exploration of the meta-reinforcement learning agent in the blasting parameter space, the blasting scheme evaluation strategy with adaptability and reliability can be efficiently generated.

The invention belongs to the technical field of air pollution monitoring, and particularly relates to an air pollution monitoring system and method based on big data. According to the method, the pollutant diffusion topological structure is constructed through collection degree fusion of the multi-source environmental data, the prediction accuracy of air pollution monitoring is effectively improved, and when the pollutant diffusion topological structure is constructed, the influence of meteorological parameters and geographic information on a pollutant transmission path is comprehensively considered, so that the prediction accuracy of the air pollution monitoring is improved. The method enables a corresponding prediction mechanism to more accurately describe a pollutant space-time evolution rule in an air environment, introduces a prediction error-based iterative optimization mechanism after a prediction result is output, and compares the pollutant prediction concentration with the actual monitoring concentration in real time, thereby achieving the real-time prediction of the pollutant. And a diffusion weight coefficient and a diffusion probability matrix in the pollutant diffusion topological structure are adjusted, so that a corresponding prediction mechanism can maintain high-precision prediction capability, and relatively accurate data support is provided for early warning of an air environment.

The invention relates to the technical field of terrain wind resource assessment, in particular to a complex terrain wind resource assessment method and system based on CFD enhancement, and the method comprises the steps: processing terrain, weather and vegetation data, and generating a complex terrain feature vector; constructing bidirectional interaction between a mesoscale meteorological mode and computational fluid dynamics, and iteratively optimizing a flow field; fluid dynamic parameters are calculated through reinforcement learning automatic optimization; turbulence simulation is enhanced by combining the generative adversarial network with large eddy simulation; dynamically allocating resolution according to the grid priority of the multi-feature calculation; the lightweight model pre-judges errors, and after the errors reach the standard, evaluation indexes are calculated to complete wind resource evaluation; the system comprises a multi-source data fusion module, a dynamic coupling module, a parameter optimization module, a reinforced turbulence simulation module, an adaptive grid module and an error prediction and evaluation module, and all the modules cooperate to achieve full-process evaluation. The problems that in the prior art, one-way coupling precision is low, manual parameter selection is low in efficiency, turbulence simulation precision and efficiency are difficult to balance, and static grids are wasted are solved.

A knowledge graph construction method based on active learning and incremental learning comprises the following steps: S1, preprocessing data from a plurality of heterogeneous data sources, and extracting entities, relationships and attributes to form an initial knowledge network; s2, vectorizing elements in the initial knowledge network by using a knowledge graph embedding model, and performing entity alignment based on vector similarity to obtain an initial knowledge graph; s3, screening out candidate knowledge triples with high uncertainty and / or high representativeness from the initial knowledge graph by adopting an active learning strategy, and obtaining user labeling information corresponding to the candidate knowledge triples; s4, performing iterative optimization on a knowledge extraction model and / or a knowledge graph embedding model according to the user labeling information; s5, new data are fused into the optimized knowledge graph in an incremental learning mode, and knowledge conflict detection and resolution are carried out in the fusion process; and S6, circularly executing the steps S3 to S5 until the knowledge graph meets a preset quality condition.

The invention relates to the technical field of tunnel engineering and intelligent construction, in particular to a shield tunneling real-time control method based on a random forest and a particle swarm optimization algorithm. The method comprises the steps that initial tunneling parameters are generated through a parameter recommendation random forest model according to geology and tunnel geometric parameters, and model hyper-parameters are optimized through a sparrow optimization algorithm; carrying out settlement prediction by utilizing the settlement prediction random forest model; if the predicted value exceeds the limit, carrying out iterative optimization by adopting a particle swarm optimization algorithm and taking the initial parameter as a starting point, and searching a global optimal tunneling parameter combination meeting the settlement requirement; finally, the optimized parameters are issued to the shield tunneling machine to be executed, the model is continuously updated based on real-time construction data, and closed-loop control is formed. According to the method, intelligent recommendation and real-time optimization of tunneling parameters can be realized, the ground surface settlement control precision and the system response speed are improved, the dependence on artificial experience is effectively reduced, and the self-adaptive capability and the intelligent level of shield construction under complex geological conditions are enhanced.

The invention relates to the field of digital core modeling, in particular to a stomatal basalt three-dimensional digital core reconstruction method, which comprises the following steps: extracting target characteristic parameters from CT (Computed Tomography) scanning gray volume data of a stomatal basalt sample, the target characteristic parameters comprising porosity, cluster number, cluster size statistics, spatial uniformity index and simplified compactness; generating an initial three-dimensional digital core model according to the target characteristic parameters; carrying out iterative optimization on the initial three-dimensional digital core model by utilizing a self-adaptive Markov chain-Monte Carlo algorithm so as to enable cluster features of the optimized three-dimensional digital core model to approach target feature parameters; and performing curvature smoothing post-processing on the optimized three-dimensional digital core model to obtain the pore basalt three-dimensional digital core model. The model not only is highly matched with real pore basalt in the aspect of macroscopic statistical characteristics, but also shows natural and smooth curved surface characteristics in the aspect of microscopic pore boundary morphology, and provides a reliable digital basis for subsequent rock physical property analysis.

The invention relates to the technical field of cloud edge collaborative computing power scheduling, and discloses a heterogeneous computing power scheduling optimization method based on a cloud edge collaborative architecture. The method comprises the following steps: acquiring real-time computing power state data of all available computing nodes in the cloud edge collaborative architecture; performing heterogeneous type division on the computing nodes according to the real-time computing power state data to generate a three-layer computing power resource pool containing cloud computing nodes, edge computing nodes and terminal computing nodes; extracting task calculation features for the current to-be-scheduled task set, wherein the features comprise calculation intensity, data dependence and real-time requirements; constructing an initial task allocation scheme based on the matching relationship between the task calculation features and the three-layer computing power resource pool; iteratively optimizing the initial scheme by adopting a dynamic load balancing strategy to generate a final task scheduling instruction; the instructions are distributed to the corresponding computing nodes to be executed, and computing power state changes in the task execution process are continuously monitored.

The invention provides an intelligent warehouse goods posture recognition and automatic sorting method and system, and the method comprises the steps: S2, extracting the edge contour and surface features of goods from point cloud data for a preliminary value set, calculating the direction vector and shape distribution characteristics of the goods through a principal component analysis method, and obtaining a quantitative result of shape feature extraction; s6, historical sorting data and real-time sensor data are extracted from the warehousing system database according to the classification basis adapted to the complex scene, the dynamic posture change trend of the goods is predicted through a time sequence analysis method, and optimization parameters of real-time processing are obtained; and S7, the movement track and the grabbing angle of the sorting mechanical arm are adjusted through the optimization parameters subjected to real-time processing, a deep reinforcement learning algorithm is adopted to conduct iterative optimization on the sorting action sequence, and an execution scheme of sorting accuracy is determined. According to the method, the accuracy of goods posture recognition and the automatic sorting efficiency in the complex storage environment are remarkably improved.

The invention discloses a cooling tower fault tracing method based on a knowledge graph, and the method comprises the steps: collecting all kinds of operation, fault symptoms and environment condition data of a cooling tower in a distributed manner, carrying out the abnormal value processing, missing completion and normalization, and constructing a knowledge graph which takes equipment components, fault modes and symptoms as nodes and causes and effects and association as edges; a fault propagation path is generated in a knowledge graph by using a structured symptom vector, a semantic feature and a historical statistical feature are combined to dynamically evaluate a path weight, and multi-round iterative optimization and sorting are realized, so that a key path with a significant weight is identified as a final fault traceability basis, and a maintenance decision is further supported. According to the invention, the traceability accuracy of complex faults of the cooling tower and the adaptive capability of the knowledge graph are effectively improved.

The invention relates to a radial frequency-doubled sound filtering metamaterial and a topological optimization design method thereof.The radial frequency-doubled sound filtering metamaterial comprises scatterers which are arranged on a base body and distributed in a concentric circle mode, and the topological optimization design method comprises the following steps that a two-dimensional axial symmetry model is established, the two-dimensional axisymmetric model corresponds to the cross section of the radial frequency-doubled sound filtering metamaterial in the radius direction and comprises periodically arranged unit structures, and each unit structure comprises a main board and a square design domain; a Floquet periodic boundary condition is applied to parallel boundaries of a unit structure in the radius direction, other boundaries are set as free boundaries, a genetic algorithm is adopted to carry out iterative optimization on the material distribution topology of a square design domain, the optimal material distribution topology of the square design domain is obtained, and then the final optimized shape of the radial double-frequency sound filtering metamaterial is obtained. Compared with the prior art, the invention has the advantages of capability of realizing omni-directional sound wave frequency doubling filtering, high design reliability and the like.

The invention relates to the technical field of reinforcement learning intelligent scheduling, in particular to an intelligent scheduling management method for prefabricated part production resources based on reinforcement learning. The specific implementation process comprises the steps of collecting order demands, material distribution and production pedestals in real time, and mapping the order demands, the material distribution and the production pedestals into production mold load tensors; when a scheduling request is triggered, inputting the production modulo tensor into a resource arrangement network, and performing search and reasoning by using a scheduling strategy based on a multi-head attention mechanism to generate a resource scheduling matching instruction; calculating a state difference tensor, and outputting an efficiency reward signal in combination with a delivery constraint and a cost constraint; and packaging the production modulo tensor, the resource scheduling matching instruction and the reward signal into semantic interaction experience, storing the semantic interaction experience into a scheduling experience playback pool for gradient modulation, and iteratively optimizing a scheduling strategy. According to the method, learning can be carried out from a large amount of historical data by utilizing reinforcement learning, iterative optimization of the scheduling strategy is realized, the calculation time consumption of scheduling instruction generation is reduced, and the flexibility of production scheduling is improved.

The invention belongs to the technical field of geophysics. According to the truncation model space three-dimensional magnetotelluric inversion method and system, a three-dimensional inversion objective function is constructed according to an inversion initial model, the inversion objective function is optimized by adopting a Gaussian-Newton method, and a Gaussian-Newton increment equation is obtained, a sensitivity matrix in the Gaussian-Newton increment equation is a sparse matrix which is obtained after truncation is carried out by adopting a distance truncation threshold value and a sensitivity amplitude truncation threshold value, the Gaussian-Newton increment equation is converted into an unconstrained least square form, and an inversion updating direction is determined according to the least square form; and performing iterative optimization according to the initial inversion model, the inversion updating direction and the model updating step length to obtain an updated inversion model. According to the invention, units with small contribution to inversion updating are effectively cut, and the model space corresponding to each piece of data is reduced.

The invention discloses an existing subway structure stress deformation assessment method and system under urban tunnel proximity construction, and the method comprises the steps: collecting stratum physical and mechanical parameters, existing subway structure design parameters and tunnel construction key parameters of an urban tunnel proximity construction influence region, and carrying out the integration to form a three-dimensional basic database; the method comprises the following steps: arranging equipment at key monitoring points to collect and preprocess real-time data, establishing a three-dimensional geology-structure numerical model based on a database, constructing an inversion algorithm by taking the preprocessed data as a target function, dynamically inverting real-time stratum parameters, and updating the database. The method comprises the following steps: constructing a stress deformation prediction model by combining real-time stratum parameters, inputting parameters to solve stress and deformation values of structures in different construction stages, setting an early warning threshold according to safety levels and specifications, comparing and outputting early warning levels, calculating model errors regularly, and supplementing sample retraining to realize dynamic iterative optimization if the early warning levels exceed the range. And safety and stability of the existing subway structure during construction are effectively guaranteed.

The invention discloses an aviation airborne software test case intelligent generation and closed-loop optimization method, which comprises the following steps of: knowledge graph construction: analyzing a DO-178C standard document and a related field document, extracting entities and relationships defined in the DO-178C standard document and the related field document, and constructing a field knowledge graph fused with DO-178C standard knowledge; initial test case generation: based on the domain knowledge graph, combining a static analysis result of the source code of the tested airborne software, and utilizing a large language model to drive and generate an initial test case set; and closed-loop iterative optimization: executing the test case, evaluating whether the structural coverage rate reaches the standard or not, automatically identifying uncovered codes when the structural coverage rate does not reach the standard, generating a supplementary test case for iterative optimization, and outputting a final test case set until a coverage rate target corresponding to the software security level is met. According to the method, the test quality and efficiency of the aviation airborne software can be improved.

The embodiment of the invention provides a power grid natural disaster early warning method and system based on artificial intelligence, and belongs to the technical field of artificial intelligence. The method comprises the following steps of: firstly, constructing a multi-source power grid and disaster information evolution association set which comprises power grid equipment operation dynamic information and natural disaster occurrence dynamic information of a corresponding region and is associated in real time through a region identifier and annotated with an information evolution trend, and then carrying out sequential risk conduction analysis on the multi-source power grid and disaster information evolution association set; then calling a pre-trained power grid disaster intelligent hierarchical prediction model to carry out multi-stage feedback prediction processing on the result, generating a power grid natural disaster hierarchical prediction result, executing early warning strategy dynamic iteration optimization according to the result, generating early warning strategy dynamic iteration adjustment parameters, and carrying out early warning strategy dynamic iteration adjustment. And finally, generating a power grid natural disaster accurate early-warning instruction based on the early-warning strategy dynamic iteration adjustment parameters, and sending the power grid natural disaster accurate early-warning instruction to a power grid monitoring center terminal. The power grid natural disaster early-warning method can comprehensively, accurately and dynamically perform early warning on the power grid natural disaster.

The invention discloses a pavement maintenance decision-making method, system and equipment, a medium and a product, and relates to the field of highway engineering management. The method comprises the following steps: firstly, collecting performance data of a target road section, and identifying a to-be-optimized pavement maintenance unit through a threshold judgment method or a K-means clustering algorithm; encoding each maintenance measure type and the corresponding maintenance opportunity into a real number vector, and taking the real number vector as a maintenance scheme code; constructing a multi-target fitness function covering pavement performance, maintenance cost and carbon emission; carrying out iterative optimization on the maintenance scheme through a particle swarm optimization algorithm on the basis, and outputting a particle swarm optimization solution set; performing rapid non-dominated sorting and congestion degree distance calculation on the particle swarm optimization solution set, and extracting a Pareto optimal solution set; and fusing the Pareto optimal solution and the full-life-cycle comparison data of the target road section, and outputting a maintenance decision scheme for each pavement maintenance unit, so that the decision efficiency and the scientificity, accuracy, sustainability and refinement degree of the maintenance decision scheme are improved.

The invention relates to a tunnel construction progress model generation method based on artificial intelligence. The method comprises the following steps: acquiring multi-source heterogeneous data such as tunnel field equipment operation and geological change, and performing multi-modal fusion to form a unified data set; based on this, a particle filter algorithm is used to correct equipment positioning deviation, and a real-time position optimization result is generated. Comparing the time and space with the progress plan, and identifying and quantifying deviation points through a critical path method and a deviation analysis algorithm to form deviation data. And constructing a construction progress model by combining construction basic parameters and adopting a genetic algorithm, generating an adjustment scheme by taking a construction period and resources as targets, and obtaining an optimal scheme through compliance screening and fitness calculation iterative optimization. By the adoption of the method, efficient integration of heterogeneous data can be achieved, the equipment positioning precision is improved to the centimeter level, multi-dimensional quantification of progress deviation is achieved, progress deviation recognition and resource allocation efficiency is improved, and a technical scheme is provided for tunnel intelligent construction.

The invention discloses a wind field rapid prediction method and system based on optimized Latin hypercube sampling and POD-BPNN, and the method comprises the steps: carrying out Latin hypercube sampling to generate an initial sample point set, introducing a sensitivity weight, and carrying out the iterative optimization of sample distribution through a greedy strategy; assembling the CFD numerical simulation flow field data of all sample points in the sample point set into a flow field snapshot matrix, and determining a dynamic multi-target truncation order and a corresponding POD mode and coefficient; building and training a BPNN model, packaging the trained BPNN model, the POD modal matrix, the mean field and the flow field reconstruction logic into an FMU module, obtaining the FMU module which can be called in a cross-platform manner, and achieving the real-time prediction of a wind field. The invention relates to the technical field of wind field prediction, significantly improves the precision and calculation efficiency of wind field prediction, and provides an efficient and accurate solution for the rapid prediction of a wind field.

The invention provides a target layout data generation method, electronic equipment and a readable storage medium. According to the method, the PCB layout generation process is converted into the iterative optimization task of the neural network model, and dual evaluation on the line length and the density distribution in the total loss function is combined, so that the problems of low layout quality, low optimization efficiency and the like in a traditional layout algorithm are effectively solved. The gradient is calculated through back propagation, the positions of the components are updated, it can be ensured that the lengths of the electrical connecting wires and the space uniformity of the components are optimized at the same time in the layout process, and therefore the overall performance and stability of layout are improved. In addition, in combination with a preset training termination condition, overtraining is avoided, and the calculation efficiency is improved. Finally, the generated optimized layout data has higher quality and consistency, and global optimization can be realized in complex design.

The invention discloses a three-dimensional scene reconstruction method and system based on monocular depth estimation, and belongs to the technical field of computer vision and three-dimensional reconstruction, and the method comprises the steps: carrying out the multi-scale feature coding of a monocular RGB image through a mixed attention depth coding module, and obtaining the hierarchical depth feature representation; carrying out autoregression depth decoding through a self-adaptive edge perception depth decoding module to generate an initial depth map; a depth confidence map is calculated through a geometric consistency constraint optimization module and is fed back to a coding module for iterative optimization, and a refined depth map is output; and three-dimensional Gaussian ellipsoid scene representation is constructed through the Gaussian ellipsoid scene reconstruction module. According to the invention, high-precision depth estimation and high-quality three-dimensional reconstruction are realized by constructing a depth-coupled closed-loop cooperative system.

The invention provides a three-dimensional wind field prediction method and system based on multi-modal complementary fusion learning. The method comprises the following steps: S1, acquiring remote sensing observation data and numerical simulation data; s2, obtaining standardized remote sensing features and simulation features; s3, obtaining a unified scene representation; s4, splicing the unified scene representation with the to-be-predicted space-time coordinates, inputting the spliced scene representation and the to-be-predicted space-time coordinates into a physical enhancement decoder, and outputting three-dimensional wind speed vectors at the corresponding space-time coordinates; and S5, iteratively optimizing parameters of the bimodal encoder, the cross-modal attention fusion module and the physical enhancement decoder to form a closed-loop prediction model. According to the method, multi-modal data complementation and physical information deep fusion are realized, through innovating a neural network architecture and a constraint mechanism, the prediction precision under a sparse data condition is remarkably improved, the physical credibility of a result is enhanced, and a technical support is provided for intelligent development of the wind power industry.

The invention provides a multi-modal data desensitization method, system and device and a medium, and belongs to the technical field of information security. The method comprises the following steps: firstly, acquiring multi-modal original data based on preset access information and acquisition frequency, and after standardized preprocessing, positioning sensitive information by using a classification recognition algorithm and generating an analysis report. Then, according to a report, matching a strategy from a desensitization strategy library bound with the sensitivity level, adjusting a dynamic confusion algorithm parameter to desensitize, integrating and generating desensitized multi-modal data, and recording and storing a mapping relationship between the original data and the desensitized data; and if a restoration request is received, after authorization verification is passed, reverse desensitization restoration data is carried out by using the mapping relation. Besides, full-process feedback information is collected, the desensitization effect is quantitatively evaluated according to a preset index, the desensitization strategy and algorithm are iteratively optimized accordingly, and effective desensitization and safety management of multi-modal data are achieved.

The invention discloses a knowledge destruction attack method and device based on an RAG system and a medium, and relates to the technical field of internet security, and the method comprises the steps: inputting a target question and an error answer into the RAG system, and generating an initial confrontation text; performing multiple rounds of iterative optimization processing on the initial adversarial text to obtain a target adversarial text; inputting the target adversarial text into a knowledge base corresponding to the RAG system; the RAG system responds to a question demand input by a user, and retrieves and outputs a question answer corresponding to the question demand from the knowledge base; and inputting the question demand and the question answer into a large language model, so that the large language model outputs a wrong answer corresponding to the target question. The method and the device are used for solving the problems of low output result precision, poor attack effectiveness and low concealment when knowledge destruction attack is carried out based on an RAG system in the prior art, and the precision of the output result is improved under the condition that the knowledge destruction attack is effectively carried out with high concealment.

The invention relates to the technical field of information, and discloses a radar dynamic anti-interference method and system based on interference source positioning. The method comprises the following steps: acquiring current signal data and a historical signal sequence, and determining an initial deviation value; extracting a historical deviation sequence according to the initial deviation value to obtain a deviation change trend vector; calculating a trend slope and a fluctuation amplitude, and determining an adjustment trigger signal; extracting feature interaction influence according to the adjustment trigger signal, and determining a weight coefficient update increment; iterative optimization is carried out in combination with the convergence rate parameter, and an optimized convergence rate value is obtained; adjusting deviation compensation model parameters according to the convergence speed value, and outputting a deviation compensation result when the compensation residual error is lower than a preset threshold value; and carrying out positioning calculation according to a deviation compensation result, and determining a corrected positioning coordinate. The model is dynamically optimized through technologies such as a sliding window and gradient descent, the problem of positioning deviation caused by complex environment signal interference is solved, and the positioning precision and the response speed are improved.

The invention relates to the technical field of automatic driving and intelligent control, in particular to a multi-transportation-equipment cooperative motion control method based on multi-stage mixed learning and a storage medium, and the method comprises the steps: generating expert demonstration data through employing a single-vehicle motion control algorithm based on an expert rule, performing initialization training on the strategy network through online iterative supervised learning to obtain a pre-training strategy model; the multi-vehicle cooperative motion control method comprises the following steps: selecting a multi-vehicle cooperative motion model, loading parameters of the model into an Actor strategy network of multi-agent reinforcement learning, performing interactive training on a plurality of agents in a simulation environment by adopting a centralized training and decentralized execution normal form, and performing online iterative optimization on the strategy based on a composite reward function and generalized advantage estimation to obtain a multi-vehicle cooperative motion control strategy. According to the method, the complementary advantages of imitation learning and reinforcement learning are exerted, the training efficiency, the strategy performance and the collaborative operation capability and robustness of the system in a complex scene are improved, and the method can be directly applied to collaborative scheduling and control of transportation equipment groups in scenes such as surface mines and ports.