3213 results about "Data entry" patented technology

The invention provides a wafer edge defect detection method and system based on image recognition and a storage medium, and belongs to the technical field of semiconductor manufacturing, and the method comprises the steps: processing annular image data, and obtaining target image data; inputting the target image data into the edge defect prediction model to obtain a defect probability distribution diagram; based on the defect probability distribution diagram, identifying a pixel region with a probability value greater than or equal to a preset threshold value as a candidate defect region; performing feature extraction on the candidate defect region to obtain a target feature vector; inputting the target feature vector into a defect classification model, and determining the defect type and classification confidence of the candidate defect region; based on the defect type and / or the classification confidence, screening out a target defect area from the candidate defect areas; obtaining three-dimensional point cloud data of the target defect area, and calculating three-dimensional geometric parameters of the target defect area based on the three-dimensional point cloud data; and determining a detection result of the target wafer based on the defect type and the three-dimensional geometric parameters.

The invention provides an ocean subsurface thermohaline reconstruction method based on multi-scale spatial-temporal feature fusion, and solves the technical problem of poor thermohaline reconstruction precision caused by incapability of capturing deep-level spatial-temporal dependence in ocean data in the prior art. The method comprises the steps of obtaining an ocean temperature-salinity anomaly feature data set, performing preprocessing to obtain time sequence data and space sequence data, inputting a multi-scale spatial-temporal feature fusion network model to perform feature fusion to obtain a temperature-salinity anomaly value, performing reverse normalization, and superposing a climate average value to obtain a temperature-salinity reconstruction result; the multi-scale spatio-temporal feature fusion network model comprises a spatial feature extraction branch, a temporal feature extraction branch and a spatio-temporal feature fusion module. The method can be widely applied to the technical field of marine science.

The invention discloses a bearing degradation trend prediction method and system based on multi-domain feature dynamic fusion and dimensionality reduction, and the method comprises the steps: collecting full-life vibration signals of a bearing, synchronously marking three stages of health, degradation and fault, constructing multi-dimensional features such as a time domain, a multi-scale frequency domain, a time-frequency domain, and the like; evaluating the cross-stage difference of the features by using double criteria of mahalanobis distance and information entropy, and adaptively adjusting the weight to complete optimization; threshold cutting, linear proportion, Softmax or hierarchical weighting strategy empowerment are automatically selected according to data distribution, and energy is reserved through PCA for dimension reduction. And a TCN-GRU deep network model is constructed. Real-time data are input into the model to predict the degradation state, if errors exceed the limit, feature reconstruction and model retraining are triggered, and full-life-cycle high-precision high-robustness multi-stage continuous online monitoring is achieved. The method aims at solving the problems that the diagnosis precision is limited and the working condition adaptability is insufficient due to the fact that single time domain or frequency domain features are excessively depended and the features of each stage of fault evolution are difficult to comprehensively characterize.

Provided is a method for perceiving and eliminating an abnormal state of active distribution network based on data enhancement, including: acquiring, by synchrophasor measurement device, data of each node of active distribution network in target domain in real-time and transmitting to processor; inputting the acquired data into a classification model, and outputting abnormal detection and classification results in real time; and analyzing the abnormal detection and classification results, and transmitting an abnormal state eliminating instruction to a distribution terminal to eliminate the abnormal state. Wherein, hidden distribution features in node data of active distribution network are mined through dynamic clustering, a large amount of unlabeled data are clustered, a data label is generated through self-coding and label correction rule, training samples with balanced category distribution is generated through data enhancement and is used to train the classification model based on dynamic graph attention network by domain adaption method.

The invention relates to the technical field of ecological environment monitoring, and discloses a coastal wetland intelligent monitoring method and system based on artificial intelligence, and the method comprises the steps: collecting unmanned plane data, satellite remote sensing data, Internet of Things sensor data and water quality monitoring buoy data of a coastal wetland; the method comprises the following steps: processing satellite remote sensing data by adopting a wavelet threshold denoising algorithm based on an attention mechanism, calibrating Internet of Things sensor data by adopting an LSTM network, and carrying out data space-time alignment based on a space-time attention fusion model to obtain preprocessed data; inputting the preprocessed data into a Transform-ResNet hybrid model to carry out environmental change evaluation, and outputting an ecological health index; when the predicted ecological health index is lower than a threshold value, a PPO algorithm is adopted to dynamically adjust a monitoring strategy according to the early warning level, and an early warning report is pushed; the whole process is intelligent, manual intervention is greatly reduced, and support is provided for coastal wetland ecological protection.

The invention provides a road underground disease detection model based on medium inversion and deep learning. A generation method of the model comprises the following steps: 1) collecting road underground radar data by using a ground penetrating radar; 2) obtaining the distribution condition of the underground dielectric constant corresponding to the B-Scan, and constructing a B-Scan-dielectric constant distribution data set; 3) constructing a deep learning medium dielectric property inversion model, and performing training by optimizing a comprehensive loss function; 4) utilizing the trained dielectric dielectric property inversion model to perform inversion on the actually measured radar map to generate an underground dielectric property distribution map; 5) performing disease classification marking according to the underground dielectric property distribution diagram, and constructing and training a disease target identification model; and 6) inputting the to-be-detected dielectric property distribution data into the disease target identification model, and outputting a prediction result of the underground hidden disease. Compared with the prior art, the method provided by the invention can realize accurate disease identification and positioning under complex underground medium distribution, and has better accuracy advantage and stronger generalization ability.

The invention provides a scene reconstruction and semantic understanding method and system based on a visual large model, and relates to the field of computer vision and three-dimensional reconstruction, and the method comprises the steps: obtaining multi-view image data of a target scene, inputting a pre-training visual large model, and outputting a joint feature representation and attention weight matrix; generating three-dimensional coordinate values and semantic probability distribution of spatial sampling points in a three-dimensional space according to the joint feature representation, and constructing a spatial semantic field; clustering the spatial sampling points by using the attention weight matrix, performing semantic consistency enhancement, and converting the spatial sampling points into deterministic semantic tags; constructing a geometric optimization objective function, and adjusting three-dimensional coordinate values; and extracting continuous space sampling points with the same semantic tag to form an object boundary, constructing a scene topological graph, deducing a scene functional structure, and generating a navigation path. According to the method, the unification of accurate geometric reconstruction and deep semantic understanding of the scene is realized, the three-dimensional reconstruction precision and semantic analysis accuracy are improved, and reliable support is provided for intelligent navigation.

The invention discloses an electromechanical equipment distributed control system based on the Internet of Things, which belongs to the technical field of distributed control, and specifically comprises the following steps: splitting a control task into task units, registration dependencies, data entries and task levels at an edge gateway, and generating a candidate scheduling range according to registration items; a calibration heartbeat with a reference segment is issued, and a driving node executes and returns a measurement record in an idle window for calibrating hardware fingerprints and link overhead and dynamically updating matching parameters; when computing power saturation is triggered, splitting a low-level task unit into suspensible fragments, setting a de-duplication execution identifier, opening a state interface, and completing migration check; selecting a target control node and a transmission path according to a calibration result, issuing fragment and state interface information, driving a target node to continue fragments according to an execution template, and guiding a source node to release occupied resources in sequence; and checking and comparing the execution result, and when the rule is triggered, recovering to the previous stable state according to the mapping rollback table.

The invention provides a reinforced soil retaining wall risk early warning method and system based on multi-source heterogeneous monitoring data, and belongs to the technical field of building risk early warning, the method comprises the following steps: inputting first data into a prediction model to obtain second data, the first data being historical monitoring data of a reinforced soil retaining wall to be detected, and the second data being historical monitoring data of the reinforced soil retaining wall to be detected; the second data is monitoring data of the reinforced soil retaining wall to be measured at the target moment; determining a construction data type based on the weather information of the target moment, and constructing data in the second data based on the construction data type to obtain construction data; and inputting the construction data, the second data and the weather information at the target moment into a risk early warning model to obtain a risk early warning result of the to-be-detected reinforced soil retaining wall at the target moment. According to the reinforced soil retaining wall risk early warning method and system based on the multi-source heterogeneous monitoring data, the accuracy and reliability of reinforced soil retaining wall risk early warning can be improved.

The invention discloses an optical surface defect data detection method based on deep learning, and relates to the technical field of optical defect detection, and the method comprises the following steps: constructing an optical scattering physical model, inputting a collected optical surface image into the optical scattering physical model for multi-modal data synthesis, and generating multi-modal image data; constructing a deep learning feature extraction network, inputting multi-modal image data, and performing multi-scale feature fusion and enhancement through a bidirectional attention feedback mechanism to generate a deep feature map; performing spatial domain analysis on the depth feature map by using a deep learning region generation method, positioning coordinates of potential defect regions, and generating a candidate defect region coordinate set; through multi-modal data synthesis driven by an optical scattering physical model, the limitation of a single imaging mode is broken through, the scattering characteristics of defects under multi-physics field coupling are dynamically analyzed, the recognizable degree of weak defects in a complex scattering environment is enhanced, and the problem of defect missing detection is solved.

The invention discloses a deep learning prospecting prediction method and system for multi-modal geological data, and relates to the technical field of mineral resource exploration and prediction. The method comprises the following steps: acquiring multi-modal geological data, and preprocessing the multi-modal geological data to obtain preprocessed geological sensing data; constructing a prospecting prediction model based on a hybrid model architecture; the hybrid model architecture comprises a CNN branch, an RNN branch and a GNN branch, and each branch is connected with the full connection layer through splicing operation; inputting the multi-modal geological data into a trained prospecting prediction model for processing to obtain a mineralization potential prediction result; the mineralization potential prediction result comprises a mineralization type, an ore body scale and an ore body grade; and based on the mineralization potential prediction result, constructing a three-dimensional mineralization potential model of the target area by using a depth generation model for visual display. According to the invention, the precision and generalization ability of prospecting prediction can be improved.

The invention discloses a supervision method and device based on construction project quality detection and a medium, and relates to the technical field of project quality supervision, and the method comprises the steps: collecting construction site data, carrying out the preprocessing, and generating a site data set; performing real-time comparison and verification on the building information model and the field data set to obtain verification data, inputting the verification data into the building information model, performing construction consistency analysis and deviation identification, and generating a quality evaluation report; performing risk assessment and priority classification on the quality analysis data to obtain risk assessment data, and performing problem priority analysis and rectification scheme formulation on the risk assessment data to generate rectification execution data; and performing rectification effect analysis and overall quality evaluation according to the rectification execution data, and generating a project quality summary report. The real-time performance and the accuracy of construction quality monitoring are improved, the construction deviation is reduced to the maximum extent, the quality management is optimized, and the smooth proceeding of a project is guaranteed.

The invention discloses an aquaculture disease prediction method based on multi-modal data fusion, and relates to the technical field of aquaculture, and the method mainly comprises the steps: collecting aquaculture data containing structured data and unstructured text data; inputting the structured data into a TabTransform model, carrying out column embedding and modeling of a context relationship between features through a multi-layer Transform encoder, and outputting a structured feature vector; inputting the unstructured text data into a pre-trained BERT encoder, and extracting an output vector marked by the CLS as a text semantic feature; and splicing the structured feature vector and the text semantic feature into a fusion feature, inputting the fusion feature into a full connection layer, and predicting the incidence probability of each target disease through a Sigmoid function. The prediction accuracy is improved, and the problems of single prediction dimension and weak generalization ability in the prior art are effectively solved.

The embodiment of the invention provides a wind field correction method based on a PPWNet model, and is applied to the technical field of wind field correction. The method comprises the steps of obtaining high-resolution grid wind field data, low-resolution grid wind field data, three-dimensional geographic coordinates and neighborhood data of a target prediction point and wind field data actually observed by a sounding point; wherein the neighborhood data comprises high-resolution grid wind field data and three-dimensional geographic coordinates of K nearest neighbor points around the target prediction point, and K is a positive integer; and inputting the acquired data into a pre-trained PPWNet model, and outputting three-dimensional corrected wind field data covering a target prediction point and a wind field divergence deviation index. In this way, the problem of data fusion of different sources, resolutions and structures can be solved, it is ensured that the output wind field is not only accurate in numerical value but also reasonable in physics, prediction results violating physical intuition are reduced, and wind field data are closer to the real situation.

The invention discloses a multi-modal data fused remote sensing image cloud classification method, system and device and a medium, and the method comprises the steps: obtaining multi-modal data and a known cloud type label, carrying out the preprocessing of the multi-modal data, forming input data, and dividing the input data and the cloud type label into a training set, a verification set and a test set according to a proportion; the wavelet consistency alignment module executes multi-scale wavelet transformation on the input data to obtain a wavelet feature tensor of each modal; constructing a cloud classification backbone network model; training and evaluating the cloud classification backbone network model by adopting the training set, the verification set and the test set to obtain an optimized cloud classification backbone network model, and inputting the optical remote sensing cloud image to be classified, the cloud profile radar reflectivity data and the atmospheric physical attribute data into the cloud classification backbone network model. And predicting the optical remote sensing cloud image, and outputting a cloud classification result label map corresponding to each pixel point in the target area. According to the invention, the cloud identification precision and stability in a complex scene are improved.

The invention relates to the field of discrete manufacturing intelligence, in particular to an intelligent decision-making and business collaboration method based on an OAG ontology and an LLM large model, which comprises the following steps: collecting historical data and business documents, and performing directional fine tuning on a basic large language model to form the LLM large model; the method comprises the following steps: analyzing discrete manufacturing scene original data and business documents, and constructing an OAG ontology library; receiving field data in real time through the LLM large model, updating the OAG ontology library, and performing feedback optimization on the LLM large model to form bidirectional feedback; integrating real-time data and historical data, inputting the data into an LLM large model, converting the data into business knowledge through layering of an OAG ontology library, and generating a main and standby decision scheme; and establishing an agent federated center, issuing a decision scheme, collecting execution data and feeding back an LLM large model, and realizing cross-scene collaboration and full-process data closed loop. According to the invention, through a technical path of a whole-process data closed loop, a core pain point that an existing LLM does not understand a business is effectively solved, and whole-link value conversion of discrete manufacturing data is realized.

The invention relates to the field of agricultural product cold-chain storage, and discloses an intelligent management monitoring system for agricultural product cold-chain storage and transportation, which comprises the following steps: synchronously sensing operation states and operation behaviors of agricultural product storage units and cold source equipment to form a data input sequence with operation association semantics; introducing a dynamic sampling mechanism based on operation event triggering into the data input sequence, and constructing a continuous evolution feature set reflecting local cold chain stability; performing time sequence analysis on a temperature and humidity parameter recovery track and a cooling capacity response rhythm presented in the continuous evolution characteristic set; delimiting a storage area corresponding to the risk evolution fragment as a key monitoring unit, and constructing an operation influence association model for describing a local cold chain constraint relationship; and performing linkage adjustment on the sensing node acquisition strategy, the monitoring update frequency and the cold source scheduling parameters under the constraint relation limited by the operation influence correlation model. The method has the advantage that the controllability of the cold chain operation process is improved.

The invention discloses a high-precision ink path bubble detection and rapid circulation compensation method and a related device. The method comprises the following steps: inputting visual sensor data, ultrasonic data and pressure fluctuation data into a trained ink path bubble semantic segmentation network model for segmentation; calculating the equivalent diameter, the bubble volume, the bubble density, the bubble flow velocity and the category of the bubbles according to the binary mask sequence; inputting the equivalent diameter, the bubble volume, the bubble density, the bubble flow velocity and the category of the bubbles into the trained multi-agent collaborative decision network model, and outputting dynamic cyclic compensation and prevention actions; and according to the dynamic circulation compensation and prevention action, the ink path pumping pressure, the valve opening degree and the ink temperature are adjusted in real time, and a high-frequency micro-pulse exhausting, reverse suction or vibration bubble removing strategy is triggered. Through high-precision detection and intelligent compensation, bubbles are effectively removed, nozzle blockage and cleaning frequency are reduced, and therefore the downtime of equipment is shortened, and waste of ink and consumables is reduced.

The invention discloses an intelligent supervision method based on a BIM technology, and relates to the technical field of constructional engineering, and the method comprises the following steps: obtaining a BIM model corresponding to the constructional engineering, constructing a digital twinning mapping relation, and carrying out the construction of a digital twinning mapping relation based on a project type, construction complexity and historical supervision data; initializing a data entry time limit through an algorithm, and changing a synchronous threshold value and a quality parameter benchmark; and collecting real-time data, wherein the real-time data comprises process progress data, material entering data, design change information and environmental safety data of the construction site. According to the intelligent supervision method based on the BIM technology, digital and intelligent transformation of supervision work is realized through deep fusion of the BIM technology and multi-source data acquisition, the matching degree of data and an actual construction scene is improved, the accuracy of anomaly recognition is enhanced through a dynamic weight distribution strategy, and the accuracy of anomaly recognition is improved. The problem of dependence on fixed threshold judgment is effectively solved, and the manual intervention cost is reduced.

The invention discloses a medical image processing method and system based on spatial adaptive feature fusion, and relates to the technical field of medical image analysis, and the method comprises the steps: obtaining original pixel data of a medical image, carrying out the preprocessing of the original pixel data, generating partition normalized volume data, performing offset correction on the partitioned normalized volume data by using a bilinear interpolation algorithm to obtain standardized medical image data; inputting the standardized medical image data into an improved double-branch feature extraction model, and respectively extracting a local feature map and a global feature map; respectively carrying out feature collaboration on the local feature map and the global feature map through a cross-branch distillation algorithm; and fusing the local feature map and the global feature map after collaboration based on a spatial adaptive fusion algorithm to generate a fused feature map, and carrying out separation convolution on the fused feature map through a gating network to generate a spatial weight map. The sensitivity of the kit breaks through a clinical threshold value, and the form specificity detection rate is greatly improved.

The invention discloses a spatio-temporal data compression method and system based on lightweight processing, and relates to the technical field of image data processing, and the method comprises the steps: based on spatio-temporal data to be compressed, establishing standardized data entries, forming a window set through time axis segmentation, carrying out the periodic discrimination of windows according to the frequency domain energy distribution, and constructing spatio-temporal blocks, establishing an error control parameter table based on the global error budget; generating a time coding stream, performing integerization on the three-dimensional coordinates and the attribute values, generating a space coding stream through double difference and run length coding, performing edge folding lightweight and texture compression on the grid data, and generating a joint coding result; and based on a joint coding result, dividing the compressed data into a base layer and a plurality of enhancement layers, and establishing a relevance hierarchical storage structure and a block-level index table to generate the compressed data capable of being transmitted in a streaming manner. According to the method, collaborative compression can be carried out by utilizing spatial-temporal data internal relevance.

The invention discloses a flood type landslide disaster monitoring and early warning method, and belongs to the technical field of geological disaster prediction. The method comprises the following steps: step 1, acquiring a flood type landslide disaster case in which a rainstorm event and a landslide event coincide in time and space, collecting landslide factor data, environment data, monitoring data and historical landslide data, and constructing a historical database; step 2, constructing and training a Bayesian network model based on a historical database; step 3, acquiring landslide monitoring data and preprocessing the landslide monitoring data; step 4, training an LSTM time sequence prediction model; 5, inputting the real-time monitoring data into the LSTM model, and predicting to obtain future landslide monitoring data; and step 6, inputting future monitoring data into the Bayesian network to obtain a slope instability probability based on a future trend. According to the invention, through organic fusion of the Bayesian network and the LSTM, dynamic prediction and real-time response of the landslide instability probability are realized, and timeliness, accuracy and robustness of early warning are significantly improved.

The invention discloses a disease prognosis evaluation and dynamic prediction method based on time sequence analysis, which comprises the following steps: acquiring disease prognosis medical data of a patient to construct a medical data time sequence set, extracting medical data time sequence characteristics and constructing a prognosis evaluation model, updating the prognosis evaluation model, and performing dynamic prediction. Inputting the to-be-evaluated patient disease prognosis medical data into the updated prognosis evaluation model to obtain a disease prognosis evaluation result, constructing a prognosis prediction model, inputting the to-be-evaluated patient disease prognosis medical data into the prognosis prediction model to obtain a disease prognosis prediction result, and updating the disease prognosis prediction result. And generating evaluation interpretation, prediction interpretation, a disease prognosis evaluation chart and a disease prediction trend chart. According to the method, the efficiency and accuracy of disease prognosis evaluation and prediction can be improved, timely and effective decision support is provided for clinic, the requirements of personalized medical treatment are met, and more accurate treatment schemes and rehabilitation guidance are provided for patients.

The invention discloses a water conservancy project construction management informatization platform, and relates to the technical field of water conservancy project management, and the platform comprises an evaluation form generation module, a form filling verification module, an intelligent approval module, a classification filing module and a safety management module. According to the invention, a whole-user, full-service and full-life-cycle construction management standardization system is constructed; automatic filling and real-time verification are realized by means of a natural language processing technology and machine learning, and data entry experience and accuracy are improved; decision support is provided through big data analysis, a filling strategy and an approval path are recommended through a model, and process management is optimized; digitized upgrading of form full-life-cycle management is realized, and the method is suitable for a water conservancy project construction management scene with a complex approval process; the state evolution trend of hidden danger points or major hazard sources in a future period of time is simulated, and early warning and plan generation are carried out based on the state evolution trend.

The invention discloses a multi-scale brain tumor segmentation method and system based on adaptive KAN, and a storage medium. The method comprises the following steps: preprocessing three-dimensional brain magnetic resonance imaging data; the preprocessed data are input into an encoder, the encoder comprises a plurality of levels, and each level executes convolution operation to extract local features, executes spatial KAN processing to extract spatial features and downsamples a feature map; the output of the encoder is input into a bottleneck layer, and the bottleneck layer captures a multi-scale global context by using a plurality of parallel expansion convolution branches; the output of the bottleneck layer is input into a decoder, the decoder comprises a plurality of stages, and each stage executes transposing a convolution up-sampling feature map, executes cross-scale gating processing to fuse encoder jump connection features and decoder features, and executes spatial KAN processing to optimize features; and the output of the decoder is input into the output module. According to the method, the problems of low calculation efficiency, poor tumor heterogeneity adaptation, insufficient multi-scale context capture and the like in the existing brain tumor segmentation can be effectively solved.

An entity-level privacy system receives a query directed towards a shared dataset, the shared dataset comprising one or more data entries associated with one or more distinct entities, each entity of the one or more distinct entities being identifiable by one or more unique entity identifiers. The entity-level privacy system implements an entity-level privacy constraint, the entity-level privacy constraint comprising a dynamic aggregation constraint based on the one or more unique entity identifiers. The entity-level privacy system determines that the one or more unique entity identifiers satisfy a threshold condition comprising a minimum number of entities. The entity-level privacy system enforces the entity-level privacy constraint on the query and generates an output to the query based on the entity-level privacy constraint and the dynamic aggregation constraint while maintaining entity-level privacy associated with the one or more distinct entities.

The invention relates to the technical field of engineering cost, and discloses a data entry method for engineering cost software, which comprises the following steps of: S1, importing an engineering quantity list file in an Excel format through an interface of the engineering cost software, and automatically analyzing a data structure of the engineering quantity list file; s2, based on a preset intelligent matching algorithm, performing dynamic mapping on the Excel data field and a data field of a project cost software function module; s3, adopting a data type self-adaptive rule to automatically classify and convert numerical values, texts and compound data in the Excel; s4, based on a data association generation algorithm, automatically establishing a hierarchical relationship and a dependency relationship between the engineering quantity list items; s5, executing one-button data import, and writing the processed structured data into a project cost software database in batches; according to the method, high-precision automatic mapping of the Excel field and the software module is realized through an intelligent matching algorithm and a dynamic weight adjustment mechanism based on the feature vector similarity, the matching accuracy is greatly improved, and the manual correction frequency is greatly reduced.

A set of images of a scene re received. Each image includes temporal data and spatial data relating to the scene. Based on the spatial data of each image, three-dimensional (3D) Gaussian splatting data is generated. The temporal data of each image and the 3D Gaussian splatting data are inputted to a neural network to generate spatial-temporal 3D Gaussian embeddings. Offset data based on the spatial-temporal 3D Gaussian embeddings is generated. The video of the scene is rendered based on the 3D Gaussian splatting data and the offset data, allowing for improved rendering of video of the scene.

The invention discloses a skin focus segmentation method based on Mama edge information and an attention mechanism, and the method comprises the steps: S1, obtaining a data set, and dividing the data set; s2, connecting an encoder and a decoder in a jumping connection mode, and introducing edge information and an attention mechanism to construct an SMEANet network for skin focus segmentation; training parameters are set, and model training is carried out to obtain an optimal model; and S3, inputting the to-be-detected data into the Mama-based edge information and attention mechanism skin focus segmentation model to realize segmentation, and finally outputting a segmentation result. According to the segmentation method, the encoder and the decoder are connected in a jumping connection mode, edge information and an attention mechanism are introduced, the SMEANet network for skin focus segmentation is constructed, the accuracy of skin focus segmentation is improved, and therefore a more reliable basis is provided for clinical diagnosis.

The invention provides a multi-source heterogeneous data processing method in the field of medical apparatuses and instruments based on a multi-modal large model, and relates to the field of intelligent monitoring of medical apparatuses and instruments, and the method comprises the steps: collecting apparatus images, audio and parameter data, inputting the data into the multi-modal large model, and fusing to generate multi-dimensional feature representation; constructing a component operation mode library for state matching; analyzing a state evolution chain to identify an abnormal propagation path and locate a fault; calculating task migration cost to generate a scheduling scheme; and lossless conversion of the system is realized. The medical instrument fault diagnosis accuracy is effectively improved, the system interruption risk is reduced, and the clinical application safety is guaranteed.