345 results about "Adaptive identification" patented technology

Provided is a self-adaptive identification method for nonlinear dynamic parameters of a reducer, which belongs to the design field of a reducer. The method includes: modeling a harmonic reducer corresponding to a flexible joint as a concatemer of a rigid reducer and an elastic torsion spring, and carrying out dynamic theoretical modeling and parameter variable independence processing on the concatemer to form a dynamic equation for parameter identification; giving an optimized motion trajectory to each joint of a robot and controlling the robot to act accordingly, acquiring relevant data needed for parameter identification based on a built-in torque sensor and double encoders inside the joint; using an offline identification algorithm to accurately identify a plurality of dynamic parameters of a collaborative robot considering joint flexibility and friction, and obtaining a minimum parameter set.

The embodiment of the invention provides a multi-factor security authentication method and device based on AI adaptive recognition and a block chain, and the method comprises the steps: collecting user biological features, behavior data and equipment information, and extracting static and dynamic features through employing a convolutional neural network and a long-short term memory network; a feature fusion network is innovatively constructed, an attention mechanism is introduced to carry out dynamic weight distribution and fusion on different features, and feature correction is carried out based on a user behavior baseline model. And meanwhile, a block chain network and an intelligent contract are deployed, so that credible storage and risk level evaluation of authentication data are realized. According to the method, the limitation of traditional single authentication is broken through, the effective balance between safety and user experience is realized, and an advanced technical scheme is provided for the field of identity authentication.

The invention provides an urban drainage pipe network monitoring data cleaning and intelligent prediction method, and the method comprises the steps: firstly obtaining pipe network monitoring data, and carrying out the classification tracking and repairing of missing values; adopting a dynamic IQR algorithm based on a sliding window to adaptively identify abnormal candidate points; secondly, introducing a pipe network topological relation, comparing upstream and downstream data change trends, eliminating non-physical anomalies caused by equipment faults, and reserving real hydraulic events; calculating the physical delay time between the nodes by using the cross correlation coefficient; and finally, constructing a random forest model, taking upstream historical data after delay alignment as feature input, and realizing accurate prediction of a future water level and quantification of a feature contribution degree. According to the method, a physical mechanism and machine learning are fused, the problems that data cleaning lacks adaptivity and a deep learning model lacks interpretability are effectively solved, and the accuracy of waterlogging early warning is improved.

The invention relates to a boiler combustion optimization control system of a thermal power plant, relates to the technical field of thermal power generation control, and aims to solve the problems of inaccurate multi-source data perception, slow control response and insufficient multi-target collaborative optimization capability under fire coal quality fluctuation and load change. The system comprises a data perception and fusion module, a working condition self-adaptive identification module, a multi-target dynamic optimization decision module and a distributed execution control module, and through multi-source data acquisition and feature extraction, depth time sequence mode identification, multi-target optimization model solving and distributed coordination control, the multi-target dynamic optimization decision module and the distributed execution control module are subjected to multi-source data fusion. Dynamic balance of boiler heat efficiency improvement and nitrogen oxide emission reduction is achieved, and adaptability and control precision of the system under complex working conditions are enhanced.

The invention relates to the technical field of remote sensing image processing, and discloses a remote sensing image adaptive identification method and system for territorial space planning, and the method comprises the steps: obtaining a multi-source remote sensing image data set of a research region, feature extraction, cloud detection, quality evaluation and adaptive preprocessing are carried out; carrying out prototype network coding, calculating a category prototype and probability, and supporting fine tuning of a set; carrying out multi-scale cavity convolution and category scale attention fusion; evaluating the adaptability score of the comprehensive fusion feature map set, and carrying out weighted fusion, classification and normalization; change detection is carried out, stable and change regions are segmented, and time sequence context features are extracted and constrained optimization is carried out; entropy is fused, a boundary is decided, uncertainty is estimated, and weighted fusion is carried out according to a change area; conditional random field optimization, confidence level grading and connected domain identification are carried out; the automation level, the adaptive capacity and the recognition reliability of remote sensing monitoring of territorial space planning are improved.

The invention discloses a power distribution network fault identification driving accurate isolation method based on an edge computing architecture, and relates to the technical field of power system automation, and the method comprises the following steps: S001, collecting voltage and current signals at each edge computing node of a power distribution network, analyzing the signal fluctuation amplitude and propagation time delay of different nodes based on the same fault event, and obtaining a fault event; and calculating a fault identification judgment difference degree between the nodes. According to the method, the identification credibility is quantified and abnormal node judgment is corrected through fault identification judgment difference analysis and topological constraint modeling; self-adaptive identification logic optimization is realized in combination with historical and real-time data; control priority and pre-simulation analysis are introduced, so that the feasibility and safety of the isolation action are improved; and finally, a closed-loop mechanism integrating recognition, simulation and feedback is constructed, and the fault response accuracy, coordination and operation stability of the power distribution network under a distributed architecture are remarkably improved.

The invention belongs to the technical field of vehicle dynamics, and relates to an automatic driving vehicle dynamics parameter adaptive identification method based on SQP and GRNN, and the method comprises four parts: automatic driving vehicle mechanism model establishment, GRNN model establishment, training set acquisition based on the SQP algorithm, and parameter adaptive identification based on the trained GRNN model. Kinetic model key parameters which are difficult to obtain by a conventional method are solved in an off-line mode by using an automatic driving automobile mechanism model and an SQP algorithm, and a solving result is used as a training set of a GRNN (Generalized Recurrent Neural Network) model; and the trained GRNN model has the capability of processing the nonlinear relationship among the steering wheel angle, the vehicle speed, the tire cornering stiffness, the suspension roll stiffness and the roll damping in the automatic driving vehicle mechanism model. Finally, accurate kinetic parameters are identified in real time according to the operation condition of the vehicle, and a foundation is laid for effective control of the automatic driving vehicle.

The invention belongs to the technical field of hypersonic flight vehicle control, and relates to a rigid-elastic coupling-oriented active and passive integrated control method for a hypersonic flight vehicle. The invention aims to realize stable tracking control of the elastic hypersonic flight vehicle. The method comprises the following steps: constructing a longitudinal dynamic model of the elastic hypersonic aircraft; self-adaptive identification of the elastic vibration frequency is realized through a cascaded self-adaptive filter; an elastic modal filtering estimation method is designed, and high-precision and low-cost elastic modal state quantity is provided for subsequent active feedback controller design; and then rigid-elastic coupling model decomposition is carried out, the control performance is ensured by using active disturbance rejection passive control for a rigid body subsystem, an RBF neural network is introduced for an elastic subsystem, an elastic mode is actively inhibited by using sliding mode control, and stable tracking of a reference instruction is realized. The method is an active and passive integrated control method for the hypersonic flight vehicle oriented to rigid-elastic coupling, and the application prospect is wide.

The invention discloses a fault remote diagnosis and digital process maintenance method for water treatment equipment, and relates to the field of water treatment equipment operation and maintenance. The method comprises the steps of unifying time and semantics, generating a quality marking time sequence, and performing working condition benchmark and tracing; according to self-adaptive identification mechanism parameters, constructing a digital twinning and trusted domain, and comparing to obtain a residual error and anomaly in real time; forming root cause evidence and sorting under topology and energy and material closed constraints; generating scripts and rollback, gray scale issuing, sentry monitoring and transaction records through steady / transient simulation and constraint checking; and convergence, energy consumption, chemical dosing, water quality evaluation, precipitation versioning template recharging and closed-loop updating are carried out. Through unified working condition data and digital twinning mechanism-data fusion diagnosis, simulation-driven gray reversible execution and knowledge template recharging under quantitative evaluation, a traceable, reusable and migratable full-link remote diagnosis and process maintenance closed loop is constructed.

The invention belongs to the technical field of self-driving automobile chassis control, and provides a lane changing scene transverse and longitudinal decoupling control method based on parameter self-adaption and application, and the method comprises the steps: firstly generating a reasonable lane changing track based on a lane changing track generation process of hierarchical planning in combination with a path planning structure; based on a genetic optimization and neural network self-adaptive trajectory tracking process, optimal lane changing control parameters are generated through test data optimization, and finally, a control parameter model is trained in combination with a neural network algorithm, so that the optimal control process aiming at a specific lane changing working condition is realized. According to the invention, high-real-time, high-comfort and high-stability driving tasks are completed through self-adaptive identification of the lane changing scene, and self control behaviors are adjusted by combining real-time identification of working conditions in the lane changing process, so that high-performance control effect output is realized.

The invention belongs to the technical field of server detection, and discloses a server PCB automatic detection system based on operation data acquisition and analysis. The system is composed of an operation data distributed acquisition module, a preprocessing and synchronous time sequence correction module, a signal integrity real-time detection module, a thermal field dynamic distribution analysis module, a time sequence drift intelligent analysis module, a multi-dimensional joint feature extraction module and an abnormal mode adaptive identification module. A fault trend depth prediction module; and a self-learning feedback and online optimization module. Acquisition units carrying FPGA hardware are arranged at key nodes of a PCB of a server, distributed acquisition of current, voltage, temperature and other multi-source electrical data is achieved by means of the high-speed parallel acquisition characteristic of the acquisition units, after asynchronism and noise interference of the acquired data are eliminated through a preprocessing and synchronous time sequence correction module, all functional modules work cooperatively, and the multi-source electrical data are acquired. A whole-process online dynamic detection system from data acquisition to anomaly analysis is constructed, and anomaly can be quickly captured in a fault germination stage.

The invention relates to the technical field of data processing, in particular to a global unattended intelligent monitoring method based on multi-source data fusion, and the method comprises the steps: obtaining a real-time feature vector of a to-be-recognized target to obtain a global discrimination score; evaluating the category uncertainty of the target to be identified; according to the category uncertainty, a dynamic weight used for model fusion is generated in a self-adaptive mode; and obtaining a local discriminant by using a pre-trained local detail model, carrying out weighted fusion on the local discriminant and the global discriminant by using a dynamic weight to obtain a final discriminant, and determining the category of the to-be-recognized target based on the final discriminant, thereby realizing adaptive recognition of the unmanned aerial vehicle and the birds. Through the technical scheme of the invention, a nonlinear decision boundary of the unmanned aerial vehicle and birds can be effectively constructed, and the recognition accuracy of a fuzzy target in a complex scene is remarkably improved.

The invention discloses a self-adaptive grabbing force control system and method of a stamping line blanking manipulator, and relates to the technical field of self-adaptive control, and the method comprises the following steps: extracting a plurality of protection layer types corresponding to a plurality of process nodes; when a grabbing instruction is received, the protection layer type corresponding to the procedure node where the target workpiece is located is determined, and a mechanical arm grabbing strategy is called; the method comprises the steps of collecting a surface image of a target workpiece, recognizing the surface image, collecting protective layer characteristics of the target workpiece, inputting the protective layer characteristics into a manipulator grabbing force control model for self-adaptive recognition, outputting grabbing force control instruction parameters, and controlling a manipulator to execute a grabbing action and carry out force control feedback adjustment. According to the mechanical arm, the technical problems that in the prior art, when stamping workpieces with different protection layers are subjected to precise recognition and matched with proper grabbing force, grabbing is unstable, the workpieces are prone to being damaged or clamping fails are solved, and the technical effect that the mechanical arm precisely recognizes the workpieces with the different protection layers and dynamically controls and adjusts force is achieved.

The invention discloses an unmanned aerial vehicle target detection method and device based on scale adaptive recognition and a medium, and relates to the field of machine learning, and the method comprises the steps: carrying out the data collection based on an unmanned aerial vehicle, and obtaining video data corresponding to a target scene; analyzing based on the video data to obtain image data of a continuous sequence, and labeling the image data to generate a training data set; performing data enhancement on the image data in the training data set; and training the convolutional neural network through the training data set after data enhancement, and performing target detection through the trained convolutional neural network. The image is used as input, and then the feature extraction fusion part is used for feature extraction. A multi-scale convolution strategy is adopted, convolution operations of various different sizes are specifically covered, and feature fusion is carried out on feature maps of different scales to obtain a feature combination.

The invention discloses a bus peak load prediction method fusing transfer learning and peak load adaptive identification, and the method comprises the steps: constructing a dynamic weight-driven transfer learning framework, and achieving the precise migration of multi-source bus knowledge to a target bus; introducing a self-adaptive threshold mechanism based on load volatility and a peak value proportion, and dynamically identifying peak values and conventional load time periods; and designing a customized loss function for strengthening the peak error weight and using the customized loss function for fine adjustment of the model, thereby remarkably improving the load prediction precision of the bus in a key period. The target bus prediction precision is obviously improved through multi-source bus transfer learning; a self-adaptive threshold identification mechanism realizes dynamic judgment of a peak time period sensitive to load fluctuation, and the perception capability of the model to a key load time period is enhanced; through the synergistic effect of the technical means, the overall prediction accuracy is finally improved, meanwhile, the prediction precision and stability of the peak load are remarkably improved, and the actual requirements of a power grid for high precision and high robustness of load prediction are met.

This disclosure relates to method and system for adaptive identification of suspicious activities. The method includes receiving image data captured by an image capturing device. For each of the plurality of frames, the method includes processing the frame. Processing the frame includes detecting, via a deep learning network, one or more individuals in the frame, and estimating, through the deep learning network, a pose of each individual and determining, via the deep learning network, behaviour dynamics of the one or more individuals based on the pose estimated in each of the plurality of frames. The method further includes classifying, via a 3D ResNet, the behaviour dynamics as one of suspicious and non-suspicious.

The invention discloses a thinking cellular system construction method for simulating spatio-temporal evolution of industrial chain nodes, and relates to the technical field of brain network construction. Comprising the steps of obtaining a thinking cell vector space; obtaining a thinking cell state; obtaining an adjacent topological relation; establishing a logic evolution dynamic potential function; constructing a thinking cellular system according to the thinking cellular vector space, the thinking cellular state, the adjacency topological relation and the logic evolution dynamics potential function; and identifying the observed state and the hidden state of the industrial chain key node based on the thinking cellular system to obtain an identification result. According to the method, dynamic simulation is performed on spatio-temporal evolution of the industrial chain key node, the evolution path of the industrial chain key node state is explicitly expressed, the evolution rule is adjusted in real time to cope with environmental disturbance, adaptive identification and risk prediction of the industrial chain key node state are realized, and the external risk resistance of the industrial chain is improved.

The invention discloses a land space use control compliance intelligent identification and auxiliary determination system, and relates to the technical field of digital treatment. A data management module receives multi-source heterogeneous time sequence data to construct a dynamic graph structure substrate, and a probability scheduling module predicts global potential conflict probability by using a space-time graph neural network and performs calibration in combination with a historical conflict database; the graph enhancement module positions a key conflict path and a core node by reversely updating a conflict knowledge graph, and the reasoning execution module activates a rule engine to execute accurate matching in a region to be subjected to fine detection and outputs a deterministic conflict identifier; the fusion decision module finally fuses probabilistic prediction and rule judgment results to generate a mixed conflict list, and the feedback module sends the mixed conflict list to the execution terminal and optimizes the model based on feedback. According to the method, multi-level and self-adaptive identification of the land and space purpose conflict from wide-area prediction to accurate judgment is realized, and the supervision efficiency and the decision accuracy are improved.

The invention provides a deep counterfeit multimedia identification method and system based on artificial intelligence, and relates to the technical field of artificial intelligence security. According to the method, high-precision forged content detection is realized through multi-modal feature extraction and link generation reasoning. The method specifically comprises the steps of multi-modal data acquisition and feature extraction, generated link reasoning and knowledge graph adaptation, multi-modal correlation analysis and dynamic threshold setting, forged path rule construction and decision model training, and self-adaptive identification strategy generation. According to the method, multi-modal features are fused, links are generated based on dynamic analysis of the knowledge graph, adaptive threshold judgment and targeted countering are achieved, the problems that a traditional method is insufficient in generalization ability, low in calculation efficiency and the like are solved, and the method can be widely applied to content authenticity identification in the fields of social media, public opinion and the like.

The invention relates to the technical field of charging optimization, and discloses a multi-stage charging optimization method based on adaptive identification of a battery, and the method comprises the following steps: S101, in-situ detection and battery type identification; step S102, applying a current step and collecting a voltage response; s103, predicting the change of the terminal voltage along with time; step S104, determining a safe amplitude limiting current; step S105, determining a pre-deceleration factor; and step S106, judging full power, generating a quota recovery instruction, and returning to the step S104 for redistribution. Battery individual parameters are obtained through in-situ detection and micro-pulse identification, current quotas are dynamically distributed based on a constant voltage proportion, a safe target current is determined by combining voltage and power amplitude limiting, current distribution unbalance is avoided, the bus current utilization rate is maximized, voltage overshoot is inhibited by means of a first-order following strategy and a pre-deceleration factor, and the safety of a battery is improved. And stable switching is ensured by matching with constant-voltage inlet current clamping.

The invention provides a building outer wall facing layer hollowing defect monitoring device and a monitoring method thereof, and belongs to the technical field of building outer wall facade facing layer defect monitoring. The monitoring method comprises a temperature data acquisition process, a heat flow intelligent analysis process, a hollowing defect adaptive identification process, a temperature field dynamic optimization process, an environmental interference self-learning correction process, a multi-source data fusion diagnosis process, a defect trend prediction process, a visual interaction process and a resource intelligent distribution process. All the processes of the monitoring method are executed through a temperature collecting and processing end, and aiming at the problems of insufficient precision, large environmental interference, low intelligence, prediction deficiency and low resource efficiency in building outer wall facing layer hollowing defect detection, through a temperature sensor matrix, an intelligent data processing process and a multi-source data fusion design, the building outer wall facing layer hollowing defect monitoring method is provided. And the detection precision, the automation degree and the maintenance efficiency are obviously improved.

The invention discloses a multi-element mixed enhanced intelligence-based cross-material pavement disease adaptive identification method, relates to the technical field of road intelligent operation and maintenance, and solves the technical problems that the identification rate is suddenly reduced and the false detection and omission ratio is high in a cross-material scene due to the fact that a model is developed for a single material or a general model is used in the prior art. The method comprises the following steps: S1, constructing a cross-material pavement disease image sample library; s2, constructing a pavement disease identification visual large model; s3, constructing a pavement disease pixel-level integrated enhanced intelligent recognition model; and S4, inputting a to-be-detected image to the trained pavement disease identification visual large model and the pavement disease pixel-level integrated enhanced intelligent identification model respectively, obtaining fusion output according to the adjusted fusion weight, and finally realizing pavement disease multivariate hybrid enhanced intelligent identification. The method can be simultaneously suitable for disease identification tasks in various material environments such as asphalt pavements, cement pavements and composite material pavements, and high-precision intelligent identification in a cross-material scene is realized.

The invention discloses a multi-protocol compatible intelligent test method and system for a power distribution automation terminal, and relates to the technical field of power distribution terminal test. The method comprises the following steps: acquiring basic information and historical test data of a tested terminal; automatically identifying and matching a communication protocol through a multi-protocol adaptive identification model, and generating a protocol conformance test template; according to the template, executing signal injection and collecting real-time response data, and calculating a terminal response matching degree and a protocol analysis accuracy parameter; and inputting the parameters into the test result evaluation model, and outputting a test result. Comprising an information acquisition module, a protocol identification and template construction module, a test execution and data acquisition module and a test evaluation module. According to the invention, automatic identification, standardized testing and intelligent evaluation of the multi-protocol terminal are realized, the testing efficiency, consistency and accuracy are improved, and the method is suitable for testing and diagnosis of distribution automation terminals of different models.

The invention discloses an edge AI-based battery thermal runaway prediction method during transportation of an electric vehicle, and the method comprises the steps: collecting multi-dimensional parameters of a battery pack during transportation of the electric vehicle, and carrying out the data fusion; performing adaptive identification based on the fused data to obtain a current scene type; then, endowing scene adjustment weights to the temperature gradient analysis data; performing fusion processing on the scene adjustment weight and the voltage fluctuation monitoring data, and extracting internal resistance change detection data; when the internal resistance change detection data exceeds a preset threshold value, generating an abnormal indication signal, extracting a dynamic change mode of multi-dimensional parameter fusion, updating a weight parameter, and obtaining a real-time weight vector; performing matrix operation on the fused data according to the real-time weight vector to obtain a comprehensive risk feature vector; and inputting the comprehensive risk feature vector into a machine learning model, determining a real-time risk assessment level, and obtaining a thermal runaway prediction result. The prediction precision and real-time performance of the thermal runaway risk of the battery pack in a complex environment are remarkably improved.

The invention relates to the technical field of video acquisition, and discloses an unmanned aerial vehicle video acquisition system and method combined with artificial intelligence, and the system comprises a scene distribution modeling module, a causal decoupling module, a sparse reasoning module, an evidence fusion module and a parameter adjustment module. The system achieves the comprehensive effects of scene adaptive identification and resource dynamic allocation, feature generalization learning and detection robustness enhancement, calculation load intelligent regulation and control and abnormal information isolation, multi-scale evidence reliable fusion and parameter autonomous optimization. Aiming at the problem of unreasonable resource allocation caused by the fact that a fixed detection strategy in the background technology cannot adapt to scene change, the system adopts real-time statistical modeling and scene type recognition technologies to realize matching of detection resources and scene requirements, and a traditional detection model excessively depends on training data distribution to cause insufficient generalization ability.

The invention belongs to the field of petroleum drilling, particularly relates to a downloading pulse signal judgment and pulse width adaptive identification method, system and device, and aims to solve the problem that a downloading instruction after the pulse width is changed cannot be automatically identified. The method comprises the following steps: acquiring pulse frequency data according to a set sampling period; filtering the collected pulse frequency data through a first smooth filtering window and a second smooth filtering window in sequence to obtain the average rotating speed and pulse waveform data of the turbine generator; and taking a curve corresponding to the pulse waveform data as a first curve, selecting a set number of points on the first curve, and determining the pulse width according to the distance between every two pulse wave crests. Under the working conditions of different well depths and different mud densities, the high downloading success rate is achieved, the pulse width can be increased at any time on the ground, the situation that a downhole tool is pulled out of the ground for resetting is not needed, and the drilling efficiency is greatly improved.

The invention provides a topology identification method and system for a low-voltage distribution network based on deep learning, and belongs to the technical field of power grid topology identification, and the method comprises the steps: deploying a multi-source sensing array at a key node position according to the structural characteristics and node distribution of the low-voltage distribution network, collecting first data through the multi-source sensing array, and storing the first data in a database; processing the data into third data; establishing a deep learning model for topology identification of the low-voltage power distribution network, training the deep learning model by using the third data set to obtain a trained deep learning model, and outputting a topological structure prediction result by using the trained deep learning model; a topological structure prediction result is displayed, and connection lines between the nodes are drawn according to the topological relation; and identifying the electrical quantity data and the environmental data of the line. Based on the method, the invention further provides a topology identification system of the low-voltage power distribution network based on deep learning. According to the invention, high-precision and high-adaptability identification of the low-voltage power distribution topological structure is realized.

The invention relates to the technical field of CIP cleaning, and discloses an adaptive pulse flow CIP cleaning control and energy efficiency optimization system, which comprises a data acquisition module, an intelligent decision and decision generation module, an execution and control module, a feedback and optimization module and a maintenance and management module, through a multi-mode flow field intelligent control and cleaning object self-adaption identification technology, various scenes of CIP cleaning can be comprehensively covered, conventional cleaning efficiency indexes are concerned, elements such as pipe diameter adaptability, dirt type pertinence and flow field mode adaptability are also included, the cooperation process of pulse parameters and a flow field is deeply optimized, and the cleaning efficiency is improved. The cleaning effect can be more accurately matched with real requirements of different cleaning objects, and a more scientific basis is provided for high efficiency and pertinence of industrial cleaning.

The invention discloses a method and equipment for controlling the virtual precession rotating speed of a micro-hemisphere gyroscope in a full-angle mode and a medium. The method comprises the following steps: enabling the micro-hemisphere gyroscope to be in the full-angle mode through three loops of amplitude, orthogonality and frequency tracking; applying a control force in the angle control equation to force the standing wave to self-rotate at a corresponding virtual angular velocity; in a standing wave precession state, identifying a total loop gain parameter in real time by using an adaptive identification technology; and the excitation voltage is multiplied by an external given excitation voltage to obtain a compensated excitation voltage, so that on-line compensation of the standing wave virtual procession is realized, and the accuracy and stability of the rotating speed of the virtual procession are improved.

The invention discloses a multi-format document structured data feature extraction system based on deep learning. The multi-format document structured data feature extraction system comprises the following steps: S1, preprocessing a document; carrying out format standardization processing on the input PDF report, wherein the format standardization processing comprises document denoising and rotation correction; s2, format self-adaptive identification is carried out; and automatically identifying the format type of the input document based on the existing document sample training model by applying a machine learning algorithm. The invention relates to the technical field of data processing, and has the beneficial effects that the efficiency is improved: the data processing speed can be remarkably improved by automatically identifying the document format and adopting a targeted data extraction method, and the data extraction speed is greatly improved compared with the prior art. And in combination with the OCR technology and the natural language processing technology, the data extraction accuracy is remarkably improved, the accuracy is improved to 90% or above, and the requirement for manual correction is effectively reduced.