426 results about "Data heterogeneity" patented technology

The invention discloses a directional detection method and system for a power supply shell based on a multi-source heterogeneous sensor, belongs to the field of material characteristic detection, and aims to solve the problems of single dimension, weak anti-interference performance, data isomerism and poor adaptability in traditional industrial detection. Through systematic innovation of multi-source sensor collaboration, dynamic mode control, cross-modal data fusion and self-learning optimization, a high-precision, high-efficiency and high-reliability solution is provided for the precision manufacturing field, and an optical, infrared and ultrasonic multi-modal collaborative detection framework is constructed; the optical unit realizes accurate capture of surface texture and morphology through polarized light imaging and 3D structured light scanning, the infrared thermal imaging unit analyzes the heat conduction characteristic of a material, and the ultrasonic array analyzes internal structure defects to form a surface-material-internal full-dimension detection capability; and nine types of defects such as scratches, pits, weld marks, bubbles, sink marks, material layering, stress cracking, thermal stress abnormity and material pollution are covered.

The invention provides an intelligent traffic vehicle behavior modeling method and system based on collaborative learning, and relates to the technical field of unmanned driving, and the method comprises the steps: obtaining traffic data in a target region; performing data calibration processing according to the traffic data to obtain a collaborative perception data set; a vehicle space-time trajectory function is generated according to the collaborative perception data set, and a conflict domain space-time distribution diagram is obtained; according to the vehicle space-time trajectory function and the conflict domain space-time distribution diagram, probability distribution characteristics of the vehicle driving intention and a multi-workshop competition cooperation relation matrix are extracted, and a multi-dimensional interaction characteristic vector is obtained; and behavior modeling is carried out according to the multi-dimensional interaction feature vectors, and a vehicle behavior model representing interaction behavior rules and decision logic between vehicles is obtained. According to the method, through multi-vehicle cooperative sensing data calibration and dynamic fusion of road geometric constraints, the problems of single-vehicle sensor shielding and cross-vehicle data heterogeneity are effectively solved, and the global consistency of vehicle position and motion state estimation is remarkably improved.

A system and method for a federated deep learning platform utilizing homomorphically-compressed and encrypted data. The system comprises multiple client devices, each with a local dataset, and a central server hosting a deep learning core. Client devices convert local data into codewords, which are also homomorphically encrypted. The central server processes these encrypted codewords without decryption, preserving data privacy. The platform supports at least two architectural variants: a conventional Transformer trained on codewords, and a Latent Transformer operating on latent space vectors. Both variants eliminate the need for embedding and positional encoding layers. The system aggregates encrypted model updates from clients, enabling collaborative learning while maintaining data confidentiality. Additional features comprise differential privacy implementation and adaptive federated optimization techniques. This innovative approach allows for efficient, privacy-preserving distributed learning across diverse datasets, addressing key challenges in federated learning such as data heterogeneity, non-IID distributions, and communication efficiency.

The invention discloses a high-resolution remote sensing image accurate classification system and method based on deep learning multi-modal fusion, and the method comprises the steps: S1, carrying out the preprocessing and fusion optimization of multi-modal data; S1.1, carrying out the standardization and normalization: carrying out the standardization and normalization of remote sensing data of different modals, and eliminating the influence caused by the difference between the different modals, the difference between the resolution, the spectral range and the like; for optical images, contrast may be enhanced by histogram equalization. The method aims at solving the problems of data heterogeneity, calculation efficiency, over-fitting, difficulty in labeling, real-time performance, interpretability and the like in an existing method by adopting multi-modal data optimization preprocessing, deep fusion model design, automatic labeling and semi-supervised learning, a lightweight model and hardware acceleration technology and a strategy for enhancing interpretability. Through the optimization, the system can maintain high classification precision, improve the calculation efficiency, reduce manual intervention, enhance the transparency and generalization ability of the model, and meet the actual application requirements.

The invention relates to the technical field of cloud computing application, and discloses a cloud computing-based shale fracture modeling platform, which comprises a multi-source data acquisition module, a heterogeneous data fusion module, a fracture dynamic modeling module, a parameter optimization module and a cloud verification module. The multi-source data acquisition module acquires various data and uploads the data to the cloud; the heterogeneous data fusion module processes data isomerism to generate a fusion feature matrix; the crack dynamic modeling module constructs a three-dimensional dynamic model; the parameter optimization module optimizes physical property parameters; and the cloud verification module verifies the model in multiple dimensions and manages the version. The platform solves the problems of traditional modeling in the aspects of data acquisition, fusion, modeling algorithm, parameter optimization and model verification by means of the advantages of cloud computing, can efficiently process multi-source data, accurately construct a crack model, improve the precision and reliability of the model, realize comprehensive verification and effective management of the model, and provide powerful support for efficient exploitation of shale gas.

The invention discloses an e-commerce personalized recommendation method based on cross-domain collaborative filtering, and relates to the technical field of e-commerce, and the method comprises the steps: collecting and preprocessing data, including user behavior data and commodity attribute data, of different fields of an e-commerce platform; materializing the collected data in different fields, and integrating the data into a structured e-commerce knowledge graph; and constructing a user-based cross-domain collaborative filtering model based on the behavior data of the user in the multiple fields, and predicting commodities in which the user is interested. By integrating data in different fields and constructing user portraits and commodity portraits, the problem of inaccurate recommendation caused by data heterogeneity in a traditional recommendation system is effectively solved, interests and preferences of users in different fields can be more accurately captured through cross-domain data association analysis, and the recommendation efficiency is improved. Therefore, personalized recommendation better meeting the requirements of the user is provided for the user, the recommendation accuracy and satisfaction degree are remarkably improved, and the user stickiness and the platform competitiveness are enhanced.

The invention relates to a personalized federal learning method based on self-adaptive local model initialization and double knowledge distillation, and belongs to the technical field of edge computing. The method comprises the following steps: performing local model initialization on each client according to cosine similarity between a global model and a local model to obtain a self-adaptively initialized local model; respectively performing knowledge distillation on each client according to the global model and the historical local model, and updating parameters of the local model through double knowledge distillation; and the central server receives local models uploaded by all clients participating in training, and obtains a final global model based on dual dynamic aggregation weights of data quality and client contribution degrees. According to the method, the precision and convergence speed of the model can be remarkably improved when the data heterogeneity problem is processed, and good adaptability and robustness are shown in various edge calculation scenes.

The invention discloses an animal epidemic disease cross-border transmission risk analysis method based on big data, and particularly relates to the technical field of data analysis. Through a multi-modal data fusion algorithm of deep learning, multi-source heterogeneous data is integrated in real time to form a unified data set, a cross-border propagation path model based on an infectious disease propagation model and an epidemiological theory is constructed, a propagation path of epidemic diseases is accurately simulated in combination with spatio-temporal information and propagation characteristics, and on the basis, the propagation path of the epidemic diseases is accurately simulated. A machine learning algorithm is utilized to analyze the fusion degree of cross-border propagation data between different regions, and based on a multi-source data fusion evaluation result, a high-risk region and a propagation speed are predicted in real time, so that potential cross-border propagation risks are evaluated, and targeted prevention and control intervention measures are proposed; according to the method, the problems of data isomerism, timeliness difference and fusion processing in the prior art are effectively solved, the accuracy, the real-time performance and the scientificity of epidemic disease cross-border transmission risk analysis are improved, and accurate support is provided for epidemic prevention and control decisions.

The invention discloses an intelligent garden management method and system based on big data, and belongs to the technical field of garden management. Microclimate region division is carried out on a garden through a clustering algorithm, and the region representative weight of each sensor is calculated in combination with a data heterogeneity response factor and a microtopography interference factor; weighted fusion of sensor data is realized, and a more representative environment estimation value is generated; according to the method, prediction and anomaly detection are carried out on a plant growth environment trend based on time sequence analysis, a garden regulation and control instruction is automatically generated, actual plant state feedback is introduced to realize dynamic optimization and adjustment of a weight, and the method effectively improves the accuracy of environment perception and the intelligent level of decision response, and improves the user experience. And resource misuse and plant health risks caused by information distortion are obviously reduced.

The invention discloses a federated learning client selection method based on reinforcement learning and a federated learning system. The method comprises the following steps: a client performs local detection training and collects loss, delay data and other related data; the central server divides the clients by using mean shift clustering based on data distribution of the clients, and regards each cluster as an independent agent; the intelligent agent dynamically selects a client based on the multi-dimensional state information, and optimization selection is carried out by adopting an exploration strategy; the client uploads model update after local training; and the central server carries out aggregation updating, and the intelligent agent optimizes client selection through a multi-target reward function according to a feedback adjustment strategy, maximizes a model convergence speed and balances communication and calculation overhead. By using the method and the system of the invention, under challenged environments of data heterogeneity, computing resource limitation, communication delay and the like, the client can be intelligently selected, the training efficiency of federated learning is optimized, and the performance and generalization ability of a global model are improved.

The invention discloses an agricultural data intelligent contract cross-domain sharing method based on a master-slave chain, and particularly relates to the technical field of agricultural data sharing. Edge computing nodes are deployed through a slave chain, multi-source heterogeneous agricultural data are collected, a dynamic trust level model is constructed, credibility evaluation and screening are performed on the data, and highly-credible data are synchronized to a main chain for sharing; the low-credibility data is reserved in slave chain verification; the main chain realizes transparency of a data sharing rule based on a smart contract, constructs a dynamic risk prediction model in combination with a space-time convolutional network, performs probability distribution prediction on a crop abnormal coefficient, optimizes a resource allocation scheme, and triggers hierarchical early warning by using the smart contract to realize efficient allocation and dynamic optimization of cross-regional resources; the problems of sharing obstacles, insufficient data credibility and inaccurate resource allocation caused by agricultural data isomerism in the prior art are effectively solved.

The invention relates to the technical field of medical system management, and particularly provides an outpatient medical insurance chronic disease medical data processing system based on artificial intelligence NLP. The system comprises a knowledge graph construction component which is configured to integrate multi-source heterogeneous data such as medical records, medical insurance and signs, perform cleaning, normalization and structural processing on the data through an NLP technology, and construct a chronic disease medical data knowledge graph; the diagnosis and treatment scheme matching component is configured to perform semantic analysis on the medical insurance policy document and the diagnosis and treatment guide through an NLP technology, and match a diagnosis and treatment scheme and a medical insurance reimbursement path in combination with the actual illness state and the medical insurance reimbursement record of the patient; the analysis and early warning assembly is configured to analyze the long-term health data of the chronic disease patient, identify potential health risks and development trends, and provide personalized health management suggestions and early warning prompts. According to the method, multi-source data integration and knowledge graph construction are achieved through the NLP technology, and the problem of data heterogeneity and the semantic association requirement are solved.

The invention relates to the technical field of petroleum and natural gas engineering, and discloses an oil and gas reservoir optimized mining method based on multi-modal data, which comprises the following steps: constructing an oil and gas reservoir multi-modal data acquisition system, seismic wave field data, logging interpretation data, production dynamic data, micro-seismic monitoring data, underground temperature and pressure time sequence data and shaft structure parameter data are obtained through the acquisition system; and performing space-time alignment processing on the acquired multi-modal data, establishing a unified geological coordinate system and a time reference, and eliminating data isomerism caused by different acquisition frequencies and spatial resolutions. A six-dimensional heterogeneous data acquisition system covering a seismic wave field, well logging interpretation, production dynamics, micro-seismic monitoring, an underground temperature and pressure time sequence and shaft structure parameters is constructed, so that the depiction precision of a reservoir porosity field, a permeability field, a saturation field and a pressure field is essentially improved.

Data heterogeneity has been shown to reduce the convergence speed and model accuracy in existing federated learning (FL) and distributed learning (DL) based artificial intelligence or machine learning (AI / ML) model training processes. In some embodiments, a group of devices participating in an FL or DL based AI / ML model training process is divided into clusters, e.g., based on similarities in their data class distributions, and each device is configured with a cascaded AI / ML model that includes a first AI / ML sub-model and a second AI / ML sub-model that are cascaded. For each device, one of the cascaded sub-models is a common AI / ML sub-model that is common to all devices in the group, and the other sub-model is a cluster-specific AI / ML sub-model that is common to the cluster that the device has been assigned to within the group of devices. A multi-stage training process for the cascaded AI / ML model is also provided.

The invention discloses a drug risk monitoring method based on multi-source data fusion, and relates to the technical field of data analysis. Through multi-source data fusion and processing, the problems of data isomerism and information islands are solved, a comprehensive and high-quality data basis is provided for drug risk monitoring, a weighted collaborative filtering algorithm and a Bayesian network are adopted, potential risk signals are accurately mined, real-time dynamic monitoring and early warning of drug risks are achieved, and the drug risk monitoring and early warning efficiency is improved. The timeliness and the reliability of monitoring are obviously improved; meanwhile, the real-time stream processing framework is used for carrying out incremental updating on newly-added data, dynamically adjusting a risk assessment result, and carrying out automatic early warning when a risk value exceeds a standard, so that the contradiction between a complex process and a rapid early warning demand is effectively solved, the efficiency and accuracy of drug safety supervision are improved, and powerful support is provided for guaranteeing the public drug safety.

The invention relates to the technical field of 5G-R network detection and monitoring, discloses a 5G-R network situation awareness method based on distributed monitoring and multi-source information fusion, and aims at solving the problems that a 5G-R network is large in scale, high in dynamic performance and complex in data isomerism. Multi-source data of a core network, a wireless network, special equipment, interface monitoring, detection equipment, a GIS and the like are collected in real time; according to the method, technologies such as Pearson's correlation coefficients, FP-Growth, Bayesian analysis, a time sequence point process, a Hookes theory, a Gaussian mixture model, a decision tree, S-ARIMA, Boxplot, N-sigma, iForest, regression analysis, a neural network and KL divergence are combined to realize multi-source data fusion, intelligent analysis and visual perception, including network alarm, application quality, operation and maintenance and resource management. According to the method, the comprehensiveness, the real-time performance, the fault diagnosis accuracy and the operation and maintenance efficiency of the 5G-R network are improved, and the requirements of low delay and high reliability of a railway scene are met.

The invention discloses a personalized federated learning method based on knowledge fusion distillation and a storage medium, and belongs to the technical field of personalized federated learning, and the method comprises the steps: a server carries out the pre-training and distributes a diffusion model to each client, and generates a local synthesis data set, initializing a global model as a student model of a client to perform knowledge fusion distillation training, and guiding the student model training by the teacher model optimized in the last round; and after training is completed, the client uploads student model parameters to the server for federal aggregation to update the global model, and meanwhile, the personalized feature extraction capability and reliability of the next round of teacher model are enhanced by using synthetic data, so that a closed-loop learning framework of global cooperation and local personalized collaborative optimization is formed. According to the method, the problems of client model drifting, performance attenuation and convergence rate slowing caused by data heterogeneity can be solved.

The invention discloses a truss structure wind-induced dynamic response prediction method and system based on physical enhancement. The method comprises the following steps: carrying out feature extraction and alignment fusion on input data containing condition parameters and wind speed time sequence data by utilizing a long short-term memory network and a physical enhancement attention mechanism; extracting multi-scale features from the fusion features through expansion convolution, and performing weighted aggregation on the multi-scale features; the physical priori knowledge of structural vibration is fused into position coding and a self-attention mechanism so as to carry out response prediction; and integrating physical model information of the truss structure and a dynamic control equation into a loss function, and calculating physical information residual loss so as to improve the physical interpretability of a prediction result. According to the method, data heterogeneity can be eliminated, complementary information can be fused, the multi-scale characteristic of wind-induced response is coped with, the accuracy and efficiency of wind-induced dynamic response prediction of the truss structure are effectively improved, and the physical interpretability and generalization ability are enhanced.

The invention discloses a federal large model knowledge collaborative training method supporting multi-modal heterogeneous clients, which comprises the following steps: each client receives a model initialization parameter issued by a central server, and applies adaptive differential privacy noise to independently train a heterogeneous lightweight model based on local private data; updating the model to which the noise is applied and uploading a modal identifier of the model to a central server side; after model updating and modal identification of each client are received, based on a modal perception weighted consensus fusion mechanism, knowledge of each client is fused to update a global large model; and the central server side issues the updated presentation layer parameters of the global large model to the client side for initialization of the next round of local training. According to the method, on the premise that a public data set or specific task setting is not needed, comprehensive compatibility of data isomerism, client dynamic participation, model diversity and privacy protection requirements is achieved, and the adaptability, stability and knowledge utilization efficiency of large model federation training are remarkably improved.

The invention relates to the technical field of injection molding production, in particular to an injection molding production real-time monitoring and intelligent scheduling management system based on the Internet of Things, which comprises a data acquisition module, a data fusion module, an anomaly detection module, a self-adaptive scheduling module and a feedback and learning module, the data acquisition module is used for acquiring running states, environmental parameters and logistics information of all equipment in the injection molding production process in real time; the data fusion module is used for carrying out preprocessing, standardization and time sequence alignment on data from different devices and different communication protocols; and the abnormal detection module is used for analyzing the fused data in real time by adopting an abnormal detection method based on a statistical threshold value and machine learning. In the invention, a full-process closed-loop system comprising a data acquisition module, a data fusion module, an abnormal detection module, a self-adaptive scheduling module, a feedback module, an online learning module and the like is constructed; the problems that in an existing system, data are heterogeneous, data quality is not high, anomaly detection lags behind, and scheduling decision is not intelligent are effectively solved.

A system and method for a federated deep learning platform utilizing homomorphically-compressed and encrypted data. The system comprises multiple client devices, each with a local dataset, and a central server hosting a deep learning core. Client devices convert local data into codewords, which are also homomorphically encrypted. The central server processes these encrypted codewords without decryption, preserving data privacy. The platform supports at least two architectural variants: a conventional Transformer trained on codewords, and a Latent Transformer operating on latent space vectors. Both variants eliminate the need for embedding and positional encoding layers. The system aggregates encrypted model updates from clients, enabling collaborative learning while maintaining data confidentiality. Additional features comprise differential privacy implementation and adaptive federated optimization techniques. This innovative approach allows for efficient, privacy-preserving distributed learning across diverse datasets, addressing key challenges in federated learning such as data heterogeneity, non-IID distributions, and communication efficiency.

The invention belongs to the technical field of asynchronous federated learning, and discloses a data isomerism-oriented knowledge alignment asynchronous federated learning method. According to the method, a data quality perception aggregation strategy is introduced, and a knowledge distillation mechanism based on the old degree is combined, so that a global model is subjected to balanced training on heterogeneous data of different devices, and the generalization ability of the model is improved. Meanwhile, a self-adaptive learning rate adjustment mechanism based on aggregation frequency and weight is designed, and it is ensured that contribution of different devices to the global model is more fair. According to the method, the training deviation in asynchronous federated learning is effectively relieved, the accuracy and stability of a global model are improved, and the method has a considerable application value for a real federated environment.

The invention provides a geological reservoir fracture segmentation method fusing a 3DU-Net graph attention mechanism and graph theory connectivity detection and a system thereof, and aims at solving the technical problems of multi-modal data registration, 3D U-Net graph attention feature extraction, graph theory connectivity detection and the like. The problem of insufficient geological reservoir fracture segmentation precision caused by data heterogeneity and connectivity abnormity is effectively solved, high-precision segmentation of a complex geological reservoir fracture network is realized through accurate alignment of multi-modal data and a graph attention feature extraction and graph theory connectivity detection technology based on 3D U-Net, and the method has the advantages of high precision and high efficiency. The problems of local misjudgment and abnormal connectivity caused by the fact that a traditional method only depends on amplitude or texture features are effectively solved, and therefore segmentation precision and robustness are remarkably improved.

The invention discloses a geotechnical engineering investigation and analysis method and system based on multi-source data fusion, and belongs to the technical field of geological prospecting, and the method comprises the steps: collecting multi-source heterogeneous data of drilling holes in a karst area, and generating a comprehensive feature vector of the drilling holes; according to the drilling comprehensive feature vector, establishing a graph attention network, and generating an update feature; according to the updated features, generating an exploration prediction result of geotechnical engineering; establishing a dynamic adjustment model, and generating a model weight set value of the graph attention network; adjusting the model weight of the graph attention network according to the model weight set value of the graph attention network; according to the method, the multi-source heterogeneous data of the drill holes are converted into the comprehensive feature vectors of the drill holes, so that the problems of data isomerism and information islands are solved; in combination with a dynamic graph attention network, the model weight is optimized in real time, and the accuracy and adaptivity of karst cave distribution prediction under the complex karst geological condition are remarkably improved; and meanwhile, continuous iterative optimization can be carried out on the model, and the geotechnical engineering investigation accuracy of the karst area is improved.

The invention provides a photovoltaic prediction method, which comprises the following steps: data preprocessing: aligning time sequences of historical power generation data and meteorological data through a time sequence warping strategy, constructing an equipment health index, and inputting the equipment health index into an input layer; designing a space-time embedding layer, and performing space-time coding on data transmitted by the input layer and all meteorological data branches; defining a neural network module, designing a local-global attention layer, designing a network layer, and injecting meteorological condition constraints through a physical regularization item; a joint attention fusion layer is designed, and key features are effectively integrated; an output layer is designed, and feature integration, linear transformation, activation function setting, physical regularization and final photovoltaic power prediction generation and output are completed; and offline prediction is realized through edge optimization. The invention provides a photovoltaic intelligent prediction method fusing time-space sparse attention and a time sequence neural network, and solves the problems of insufficient data isomerism, calculation efficiency, physical interpretability, dynamic environment adaptability and the like in the prior art.

The invention discloses a personalized federated learning method based on a kernel distance between clients, a framework and an application, and belongs to federated learning and an application technology thereof. The method aims at double challenges of data isomerism and privacy protection, and is based on a federated learning framework of a conditional policy network and differential privacy. The CPN adaptively balances the conflict between the global consistency and the local personalized characteristics by dynamically generating the weights of the personalized characteristics and the global characteristics, so that the influence caused by data isomerism is effectively relieved. Meanwhile, the differential privacy technology protects the sensitive data characteristics of the client and reduces the risk of privacy disclosure by injecting Gaussian noise in gradient updating. The personalized federal learning method provided by the invention not only is superior to the existing mainstream method in global model performance and fairness between clients, but also realizes good balance between privacy protection and model performance.

The invention discloses a federated learning method based on partial label mask weighted distillation, which comprises the following steps of: constructing a federated learning system comprising a central server and a plurality of clients, and broadcasting a global model to each client by the central server; the client performs calculation by using the received global model to obtain a teacher model, regards a local model as a student model, performs distillation training on the local model by using local data, and uploads trained model parameters to the central server; and the central server performs federal aggregation on the updated local model parameters, and issues the aggregated and updated local model to each client for next round of local training. According to the method, the client side learns knowledge of the global data set in the teacher model through partial label mask distillation, learning of minority class samples in the local data set is enhanced, learning of the local model on the knowledge of the global data set is enhanced by constructing the teacher model with more complete knowledge, and the learning efficiency of the local model is improved. And the risk of disastrous forgetting caused by data isomerism is reduced.

The invention relates to a medical health big data exploration system, and the system comprises a demand understanding module which is used for converting a data exploration demand into a structured exploration task; the compliance review module is used for performing compliance review on the probing task; the code generation module is used for generating a query code according to the probing task and the target database characteristics; the query execution module is used for querying query result data corresponding to the query code from a target database according to the query code; the query result analysis module is used for performing multi-dimensional analysis on the query result data to obtain a multi-dimensional analysis result; a report generation module; and the result presentation module is used for presenting the exploration report in an interactive mode in a user interface and supporting iterative improvement of the report by providing a feedback mechanism, so that the problems of data isomerism, difficulty in quality evaluation, low exploration efficiency, high professional threshold, insufficient privacy protection and the like in the existing medical data exploration process are solved.

A federated byte latent transformer platform utilizing homomorphically-compressed and encrypted byte-level data. The system integrates dynamic entropy-based patching into federated learning to enable efficient, robust, privacy-preserving collaborative learning across distributed nodes. Client devices convert local data into dynamically sized patches based on entropy thresholds, encrypt these patches, and send them to a central server that processes them without decryption. The system offers improved robustness to input noise, enhanced character-level understanding, and better adaptation to low-resource languages compared to token-based approaches. It enables simultaneous scaling of both patch size and model size while maintaining fixed inference budgets, allowing efficient deployment on resource-constrained devices. These innovations address critical challenges in federated learning: efficiency, robustness to data heterogeneity, and privacy preservation.

The invention belongs to the technical field of electric power automation, and particularly relates to an automatic modeling method of power grid dispatching knowledge based on a large model and a related system.The electric power knowledge is decomposed into an explainable intermediate reasoning path through the thinking chain technology, and adaptive small models are selected according to the path to form a modeling link; a modeling link is represented as a triple form, edges and nodes are complemented, a complete knowledge graph is obtained, problems in the knowledge graph are further decomposed to construct a reasoning link, meanwhile, input data are monitored in real time, newly added entities are associated with the reasoning link, and unified representation and processing of heterogeneous power system data are achieved. Due to the fact that data from different sources are different in format, precision and semantics, according to the method, through thinking chain decomposition and small model link processing, sub-module optimization and dynamic adaptation aiming at different data types are achieved, and precision loss and adaptation difficulty caused by data heterogeneity are effectively solved.