350 results about "Reachability" patented technology

The invention provides a business vitality prediction and business district evaluation method based on multi-modal feature fusion, and relates to the technical field of spatial data analysis, and the method comprises the steps: obtaining business POI data and time sequence trajectory data of a target area, extracting the score and emotion features of user comments, and calculating the pedestrian flow distribution based on the time sequence trajectory data; then inputting the features into a recurrent neural network to predict a commercial vitality time sequence, constructing POI data into a knowledge graph, and constructing and optimizing a network reachability matrix in combination with position information and human traffic; and finally, dividing a target region by using a Louvain algorithm, fusing multi-dimensional features by edge computing nodes of each sub-region, dynamically adjusting the weight of an evaluation index through reinforcement learning, and outputting a commercial service evaluation result. By fusing multi-source data and a deep learning method, the life circle business service quality can be evaluated more accurately and comprehensively, and decision support is provided for urban planning and business development.

The invention relates to a network attack tracing method and device based on a threat graph, equipment and a medium. The method comprises the steps of obtaining network security log data from multiple security data sources, and performing standardization processing to obtain a structured network security event data set; extracting threat entities and behavior relationships among the threat entities from the structured network security event data set to obtain an entity set and a relationship set; constructing a threat map according to the entity set and the relationship set, and performing time data slicing according to the timestamp to obtain a map snapshot and map metadata; based on a predefined attack chain template, identifying an attack chain sub-graph conforming to an attack behavior structure in the threat graph to obtain an attack chain set and a path reachability matrix; and according to the attack chain set and the path reachability matrix, attack path inversion is carried out by taking the target node as an end point, and an attack traceability path and a graph evolution process display result are obtained. By adopting the method, the network attack path can be identified and the attack source can be traced.

The invention provides a multi-scene access control dynamic verification method and system based on a permission tree, and relates to the technical field of scene access, and the method comprises the steps: receiving a user resource access request, obtaining a user permission tree, obtaining a mapping rule according to a scene identifier, screening to generate a primary screening permission node set, carrying out the grouping processing through employing a consistent Hash algorithm, and obtaining a primary screening permission node set; and generating a scenarized permission tree through reachability analysis, merging and reconstruction, performing hierarchical verification on identity validity, scene legality and resource access permission on the request, and generating an access control strategy. The method can dynamically adapt to authority requirements of different scenes, and the accuracy and flexibility of resource access control are improved.

The invention discloses an anomaly detection method based on global reachability density and an adaptive threshold mechanism, which comprises the following steps: collecting single-index historical water quality monitoring data of a target monitoring station, extracting a subsequence with a fixed length through a sliding window mode, constructing a neighbor index structure and extracting neighborhood information, acquiring a k neighbor set of each data point; respectively calculating the local accessibility density of each data point in the local neighborhood of the data point; respectively calculating the global reachability density of each data point; performing statistical summary on GRD values of all data points in the sample set, calculating a mean value and a standard deviation of the GRD values, and defining a dynamically adjusted distance threshold value based on a variable coefficient; and an abnormal score function fusing local and global density features is further constructed, and abnormal point determination is realized according to a distance threshold. The method is suitable for complex time sequence water quality monitoring data, can effectively improve the recognition accuracy of weak anomalies, local drifts and extreme abrupt changes, and has high universality and robustness.

The invention discloses a three-dimensional modeling method and system, and relates to the technical field of intelligent perception and three-dimensional reconstruction, and the method comprises the steps: taking BIM as a priori, sampling a geometric entity as a visibility evaluation point cloud, and constructing a graph structure environment state under the constraints of a field of view, distance measurement, an incident angle, overlapping and other sensors; on the basis, a deep reinforcement learning agent which is pre-trained by a synthetic scene and subjected to domain randomization migration is introduced, stations and scanning parameters are selected online according to an observation-decision-execution-update closed loop, visibility and coverage are re-estimated after each step of scanning, and self-adaptive correction is carried out on an unexecuted sequence in combination with online optimization; compared with an off-line global optimization method, the method has the advantages that a search space is effectively compressed through candidate station pre-generation and visibility gating, and continuous and adjustable balance is formed among coverage, point cloud quality and operation time by multi-target awards; for engineering constraints such as temporary shielding, site reachability, registration overlapping degree and the like, the strategy can be dynamically replanned in an execution period.

The invention discloses a diffusion model and explicit constraint guidance fused visual navigation method and system. The method comprises the following steps: performing feature extraction on visual observation data to obtain target observation semantic features; performing feature extraction on the navigation target data to obtain target image features; performing feature fusion on the target observation semantic feature and the target image feature to generate a condition vector; inputting the condition vectors into a condition diffusion model for diffusion training to generate a plurality of initial candidate paths; constructing an explicit constraint based on the initial image data, and performing path optimization on each initial candidate path according to the explicit constraint to obtain a plurality of candidate optimization paths; the explicit constraints comprise collision constraints and target distance constraints; and determining a target navigation path from each candidate optimization path according to a path selection strategy. According to the method, the problems of insufficient path safety and poor dynamic adaptability can be solved, the safety and accessibility of the path are remarkably improved, the path navigation efficiency is further improved, and the method can be widely applied to the technical field of computers.

The invention discloses a dynamic scheduling optimization method based on deep reinforcement learning and a heterogeneous graph neural network. The method comprises the following steps: firstly, converting a flexible job shop scheduling problem into a Markov decision process, and setting a state, an action, a state transition and a reward function; then constructing a basic heterogeneous graph and an enhanced heterogeneous graph, and defining the types and the number of nodes and edges and a directed relationship between the edges; performing three-stage feature embedding by adopting a heterogeneous graph attention network to obtain machine node embedding, operation node embedding, distribution instance node embedding and global state features; and inputting an action state vector formed after each feature is processed, optimizing a decision network according to a reward function, updating parameters to obtain a flexible job shop scheduling model, and completing FJSP solving. The method can effectively consider the flexible job shop scheduling problem under transport time and machine reachability constraints, can also process scheduling problems of different scales, has good generalization ability, and shows excellent performance in large-scale application.

The invention discloses a logistics transportation personalized recommendation path planning method and system based on a cloud platform, and belongs to the technical field of logistics transportation, and the method comprises the steps: constructing a heterogeneous traffic map model fusing road sections, geographic interest points and traffic event information; collecting historical transportation task records of a plurality of users, generating user semantic intention tags based on delivery behaviors, and mapping the user semantic intention tags to traffic map nodes and edge attributes to form a semantic constraint graph structure; performing joint modeling on the user intention and the path reachability by using a heterogeneous graph neural network to generate a path scoring sequence; after the transportation task is completed, abnormal events are collected, and the semantic intention label mapping relation is dynamically updated based on reverse alignment errors; the model and the mapping module are deployed on a cloud platform, the traffic state is obtained in real time in combination with edge equipment, and path recommendation online generation and rapid pushing are achieved; according to the method, the personalized matching degree and the real-time response capability of path recommendation can be effectively improved, and the method is suitable for intelligent logistics path planning in a complex scene.

The invention discloses an equipment disassembly sequence automatic generation and verification method based on a physical engine, which belongs to the technical field of equipment disassembly, and realizes component physical attribute and dynamic connection constraint loading by acquiring an equipment three-dimensional model and initializing a physical engine simulation environment. And in the physical engine simulation environment, iteratively evaluating the feasibility of the disassembly operation, accurately capturing physical behaviors such as gravity and collision, and identifying disassembly obstruction and preset operation to generate candidate operation. A path search algorithm is utilized to intelligently explore and generate a creative disassembly sequence, and the limitation of traditional artificial experience and a static method is overcome. Through a multi-dimensional nonlinear total cost function, factors such as the number of disassembly steps, collision risks, tool replacement and accessibility and the like are comprehensively quantified, and an optimal solution considering efficiency, safety and operation convenience is verified and selected.

The invention relates to the technical field of medical information, and discloses a knowledge graph driven intelligent analysis system based on the medical field, and the system comprises a knowledge graph construction module which is used for constructing a medical knowledge graph containing a subject, an event, an object entity and an information reachable relation; the behavior pattern learning module is used for learning a parameterized behavior decision model for the subject entity; the dynamic deduction module is used for deducing the market change after the event is injected according to the behavior decision model and the information reachable relation; and the strategy generation module is used for reversely generating a market strategy by taking the dynamic deduction process as a fitness function. The method comprises the following steps: constructing the medicine field knowledge graph; learning a behavior decision model of the subject entity based on a graph; dynamically deducing the market change in a virtual environment; and reversely solving and generating the market strategy. According to the method, prospective dynamic deduction can be carried out on the market, the optimization strategy is actively generated, and the scientificity and timeliness of decision making are improved.

The invention relates to the technical field of resource scheduling, in particular to a digital-intelligent non-accompanying service management scheduling system, which comprises a service baseline construction module for instantiating a service item into a standard plan event flow; the state portrait generation module is used for collecting service object data flow, establishing a dynamic time sequence signal by applying a recurrent neural network model, and quantifying the occurrence probability of a future abnormal event through multi-step prediction; when the triggering is abnormal, executing causal two-way collision positioning reasons by applying the knowledge graph; the task load calculation module is used for integrating the plan event flow into a plan task component, and generating a worker resource list in combination with the predictive risk score and the worker real-time load and space reachability analysis containing the fatigue index; the system further comprises a task distribution module, the optimal staff is determined and a closed-loop scheduling instruction is generated by minimizing a composite cost function containing predictive load cost, and digital intelligent unattended service management from prediction to scheduling is realized.

The invention relates to the technical field of intelligent scheduling, and particularly discloses an intelligent scheduling decision control method and system for a distributed car washing robot, and the method comprises the steps: constructing a weighted graph model which comprises a vehicle position, obstacle distribution and path reachability, and dynamically adjusting the weight of an edge according to the path length, energy consumption and time; optimizing the graph structure by adopting an improved minimum spanning tree algorithm, and calculating a path efficiency index in combination with fuzzy logic and a neural network technology; on the basis of the optimized subgraph, applying a shortest path algorithm to obtain a local optimal path, and introducing a dynamic adaptation coefficient to evaluate the response capability of the path to environment change; fusing the path efficiency index and the dynamic adaptation coefficient into a comprehensive scheduling feature vector, inputting the comprehensive scheduling feature vector into a trained deep learning scheduling model, and predicting an optimal scheduling strategy in combination with a task priority and a robot resource state; and dynamically adjusting the moving path and the operation density of the robot according to a scheduling result, and realizing self-adaptive evolution and cooperative control of the system.

The invention discloses a manufacturing task autonomous negotiation and execution method based on a large language model agent, and the method comprises the steps: constructing a production scheduling agent, an equipment management agent, a material distribution agent and a quality control agent, analyzing a natural language task instruction through the production scheduling agent, and decomposing the natural language task instruction into subtasks; each agent calculates a utility value based on the load rate, the resource matching degree, the estimated completion time and the historical success rate, and performs structured negotiation to achieve a task allocation consensus; a prediction-check-rollback architecture is adopted to generate an action instruction sequence, the sequence is compiled into a time Petri network transition sequence, and reachability verification is carried out based on hard security constraints; production environment data is collected in real time to trigger anomaly detection and re-negotiation, and a formalized security verification and causal anti-factual reasoning parameter updating mechanism is introduced. According to the method, unstructured instruction understanding, autonomous task planning, multi-agent collaborative decision and closed-loop optimization are realized, and the problems of real-time performance, safety and interpretability of a large language model in manufacturing control are solved.

The invention discloses an enterprise and policy service automatic matching method based on machine learning, and the method comprises the steps: carrying out the processing of enterprise multi-source data, constructing an enterprise portrait graph, and generating an enterprise graph embedding and structuring feature set; performing policy text analysis and condition recognition, constructing a policy condition graph and generating condition graph embedding representation; constructing a condition constraint field based on policy conditions, and generating a cross-graph alignment relationship and fusion features; integrating a multi-source feature input improved model to carry out joint modeling, and outputting a basic matching score; constructing a condition boundary manifold, and generating an anti-fact feature sample and a matching elasticity score; and based on the basic and elastic scores, generating a comprehensive score and outputting matching result information. According to the invention, by introducing graph structure perception alignment, conditional constraint guide interaction and a multi-channel scoring aggregation mechanism, high-precision, high-interpretability and intelligent reachability automatic matching between enterprises and policy services is realized.

The invention relates to a source code vulnerability slice-level detection and statement-level positioning method based on multi-task learning, and belongs to the technical field of source code detection. The invention provides a new source code vulnerability detection method based on deep learning. According to the method, the program slice sub-graph related to the vulnerability candidate elements is extracted from the whole source code based on the accessibility of the program dependency graph, and vulnerability detection is carried out by taking the slice sub-graph as granularity, so that vulnerability features can be better captured, and the detection accuracy is improved.

The invention relates to an airplane airborne BIT system testability verification method based on a correlation model, and the method comprises the steps: carrying out the testability modeling of an object system, and obtaining a dependency relation model which describes the relation between a fault mode of the object system and a test method of the object system; based on the dependency relationship model, obtaining a BIT theoretical fault report set of each fault mode under all test methods by utilizing reachability analysis, and deducing an optimal diagnosis fault fuzzy group; constructing a fault sample library, verifying the BIT theoretical fault report set by using samples in the library, and adjusting the dependency relationship model according to a verification result; and performing fault simulation based on the verified dependency matrix, and calculating a total fault detection rate and a fault isolation rate of the object system according to a generated fault report. According to the method, the isolation rate index of the fault can be effectively verified in the incomplete and unformed stage of the fault diagnosis strategy.

The invention discloses a GIS resource management method, system and device and a medium, and belongs to the technical field of resource management, and the method comprises the steps: obtaining a city planning map, constructing a city three-dimensional model, and collecting multi-source city feature information; constructing a multi-source data fusion matrix of a GIS graph and generating three layers of attribute tags; a spatial gravity model is constructed, and reachability analysis is carried out on urban resources; constructing a dynamic prediction model, and triggering a model iteration mechanism; and generating a resource health degree index based on a model iteration mechanism result, and triggering a resource allocation control signal. The method has the beneficial effects that through multi-source data fusion matrix construction, precise space-time association of geographic space, Internet of Things time sequence and visual feature data is realized, and the integrity and accuracy of data integration are improved; quantitative evaluation and prospective pre-judgment of resource supply and demand are realized based on reachability analysis of a space gravity model and a dynamic prediction model of a space-time diagram convolutional network.

The invention relates to the technical field of project site selection decision support, in particular to an intelligent project site selection decision support system and method based on ArcEngine. The method specifically comprises the following steps: integrating site selection data, and realizing unified management of geographic information and coordinate system conversion by utilizing ArcEngine; missing values and abnormal values are processed by adopting multilayer perception interpolation, Bayesian interpolation, adaptive density clustering and random forest correction technologies, and data reliability is ensured through a standard verification code mechanism; building a comprehensive evaluation model of resource accessibility, environmental suitability and cost control, generating an optimal solution and simulating a long-term industrial evolution trend; through industrial chain modeling, multi-agent game analysis and a network optimization algorithm, node benefits and a synergistic effect are quantified, and supply chain dynamic scheduling is optimized in combination with reinforcement learning. According to the method, site selection scientificity and industrial cooperation benefits are remarkably improved, and the method is suitable for complex site selection requirements of the modern coal chemical industry heavy asset industry.

The invention relates to a multi-mode network accessibility calculation method for comprehensive traffic, a calculation device and a storage medium, and the method comprises the steps: constructing a network structure of a comprehensive traffic network which comprises a road network, a railway network and an aviation network; recording longitude and latitude coordinates of a starting point and an ending point; a heuristic function of an A * algorithm is improved; the hierarchical search algorithm comprehensively considers travel paths in the traffic network, including intra-city travel paths of cities where a starting point and a terminal point are located, and inter-city travel paths connecting the cities where the starting point and the terminal point are located; calculating total travel time; and calculating reachability and diagnosing reachability depression. According to the method, the path search algorithm is improved with the aim of improving the efficiency, and the path search algorithm suitable for the multi-mode traffic network is constructed from the two aspects of optimizing the search path and improving the search strategy. And by introducing a hierarchical search strategy, unnecessary repeated search is effectively reduced, and the search efficiency of the algorithm is improved. Meanwhile, the algorithm fusion considers the influence of the transfer times and the train connection relationship on the travel path selection, and the actual travel behavior of the traveler is simulated more accurately.

The invention discloses a graph structure data reachability index construction and query method and system meeting time and step length constraints, and the method comprises the steps: obtaining tense graph data, and constructing an index based on a 2-hop label; optimizing the index by using a pruning technology, and constructing a hierarchical index based on time merging and step length merging; reading corresponding indexes according to the query, and filtering indexes which do not meet conditions according to constraints; and executing the query algorithm to obtain a query result. According to the Hop label index construction method combined with the pruning technology, tense correlation pruning is added on the basis of the traditional 2-hop index technology, and the index scale and redundant path storage are greatly reduced. The time merging and step length optimization technology is adopted, the query efficiency is remarkably improved through index merging and step length control, and meanwhile storage occupation is reduced.

The invention relates to the technical field of intelligent prediction, and discloses an emergency resource reserve intelligent prediction system and method based on a graph neural network, and the system comprises a space-time diagram construction module which collects port, berth, channel segment, emergency warehouse and ship information, passing time window information and emergency information, and constructs an event flow heterogeneous space-time diagram; the backbone screening module is used for screening backbone sub-graphs and setting a relation feature conversion rule and a time feature representation rule; the graph memory updating module is used for updating node and edge features; the multi-task prediction module is used for outputting quantile demand prediction, reachability scores, transportation timeliness and safety stock threshold values; the consistency constraint training module is used for executing consistency constraint training; the inventory allocation decision-making module is used for generating allocation distribution quantity and outputting predicted arrival time; and the berth time window recharging module is used for generating a berth operation time window and recharging the actual arrival time. According to the invention, emergency resource demand prediction, inventory threshold calculation and allocation and berth operation linkage decision are realized.

The invention relates to a mechanical arm pose reachability enhancing method based on a rotary table and a related device, and the method comprises the steps: setting a mechanical arm working range space on a simulation platform, randomly generating the pose of a viewpoint in the working range space, detecting whether the mechanical arm can normally reach the generated viewpoint pose or not, and if yes, determining whether the mechanical arm can normally reach the generated viewpoint pose. Recording the pose of each viewpoint and the arrival capability result of the corresponding mechanical arm according to the detection result; training a pose reachability evaluation and attention fusion network by using the constructed simulation data set, taking the trained pose reachability evaluation and attention fusion network as a classifier, and incorporating the trained pose reachability evaluation and attention fusion network into a pose reachability enhancement generative adversarial network to evaluate the reachability of the pose generated by a generator; and after the training is completed, carrying out online deployment on the pose generated by the generator by using the pose reachability enhanced generative adversarial network. According to the method, the accessibility of the mechanical arm under the given attitude can be accurately evaluated, and the mechanical arm can flexibly and accurately execute tasks by generating the rotating angle of the rotary table.

The invention discloses a public facility reachability evaluation method and system based on geographic multiple agents, and belongs to the field of urban calculation and geographic information. According to the method, a multi-agent system comprising a user agent and a facility agent is firstly constructed, and then a dynamic utility function combined with an evolutionary game theory is introduced to simulate the dynamic decision and congestion avoidance behavior of a user; and user strategies are balanced through multiple rounds of iterative evolution, so that a macroscopic reachability pattern emerging from a microcosmic game is obtained. According to the method, starting from the microscopic individual level, by simulating the dynamic game process of a large number of agents, the defects of a static calculation model are overcome, and dynamic evaluation and predictive simulation of reachability of public facilities are achieved. Compared with the prior art, the evaluation result is more accurate, dynamic feedback of the real world can be reflected, and a scientific decision support tool is provided for urban planning.

The invention discloses a dynamic order distribution optimization system based on deep learning, and relates to the technical field of order management, and the system comprises a multi-dimensional dynamic data collection module which is used for obtaining the multi-dimensional dynamic data of each service item in real time; the project emergency feature calculation module generates a project emergency feature value reflecting the resource allocation priority; the resource reachability feature extraction module outputs a resource reachability feature value reflecting the task assignment difficulty level; the composite dispatch risk feature fusion and optimization model module performs dynamic weighted fusion on the two types of feature values to generate a composite dispatch risk feature vector, inputs the composite dispatch risk feature vector into an asynchronous parallel genetic optimization model to perform dispatch path search, and introduces a convergence premature recognition mechanism to prevent the algorithm from falling into local optimum; and the dispatching strategy dynamic adjustment and feedback optimization module continuously optimizes model parameters through feedback of an actual service execution result to realize closed-loop iteration update of a dispatching scheme.

The invention discloses an asset vulnerability repair priority evaluation method based on an attack graph, and the method comprises the following steps: S1, carrying out the detection of all assets, and carrying out the vulnerability scanning of the detected assets; s2, obtaining a network access relation of the detected assets; s3, constructing an asset access relation graph based on the network access relation; s4, extracting external reachable vertexes based on the asset access relation graph; s5, carrying out asset importance calculation on the extracted vertexes; and S6, performing vulnerability priority ranking according to asset importance. According to the method, the accessibility of the network access relation is considered when the vulnerabilities are subjected to priority ranking, various attack paths possibly existing in the network can be fully reflected, and therefore the asset vulnerability repair priority can be more comprehensively and accurately reflected.

The invention discloses a data processing method and device of a data center network, equipment, a medium and a product, and relates to the technical field of data processing, and the method comprises the steps: obtaining network topology information and a data forwarding table rule corresponding to each equipment subnet in the data center network; performing parallel coding on each piece of network topology information and each data forwarding table rule based on a network parallel binary decision diagram to obtain a data forwarding path structure between subnets on each piece of equipment, the network parallel binary decision diagram being generated based on a lockless index and hierarchical cache; based on the path connectivity corresponding to the data forwarding path structure, constructing a data forwarding graph corresponding to the subnet on each device; and verifying the data reachability between the subnets on each device according to the data forwarding graph, and generating a data reachability verification result. According to the technical scheme, the concurrency limitation is overcome, high concurrency can be achieved, and the data plane verification efficiency of the data center network is improved.

The invention relates to the technical field of computers, in particular to a physical space script game delivery system and method supporting rapid deployment, and the method comprises the steps: receiving a request of a user side, and generating a space coordinate graph; constructing a spatial feature set containing function / topological attribute vectors and a task demand set containing function demand vectors, difficulty coefficients and the like; calculating a matching degree and screening candidate point locations; constructing a task logic chain and an accessibility matrix, and performing recursive verification to obtain a binding scheme; a playing process and a two-dimensional code are generated and issued to a code scanning tourist end; according to the method, the operation investment can be effectively reduced, the deployment period is shortened, the coordinate graph generation of any scene is supported, the dependence on specific field hardware is eliminated through standardized vector construction and logic matching, the problem that a traditional scheme is difficult to reuse is solved, and the method can adapt to any physical space to realize the rapid deployment of the script game.

The embodiment of the invention discloses a reachable probability calculation model training method, a control method, equipment and a storage medium. The training method comprises the following steps that according to design parameters of the mechanical arm, to-be-screened tail end pose data are determined; selecting reachable end pose data from the end pose data to be screened, and generating a training set; a multivariate Gaussian mixture model used for fitting reachable pose data space distribution is obtained based on training of the training set, a reachable probability calculation model is generated based on the multivariate Gaussian mixture model, and the reachable probability calculation model is used for outputting the reachable probability corresponding to any tail end pose. Therefore, the calculation model can be constructed and trained by adopting the architecture of the multivariate Gaussian mixture model. When control is carried out, control is carried out through model calculation of the reachable probability, the failure caused by the fact that the mechanical arm can be grabbed but cannot be reachable is avoided, and the operation success rate of the mechanical arm in the complex environment is increased.

The invention relates to the technical field of virtual training, in particular to a firefighter search, rescue and escape virtual training method and system, and the system comprises a virtual scene dynamic construction module which abstracts a room and a corridor intersection into nodes according to a planar graph of a building, abstracts a door and a channel into edges with geometric distance weights, and carries out the virtual scene dynamic construction module; and responding to a fire spreading or structure collapse event in the virtual training, adjusting the weight of a corresponding edge to be infinite, obtaining a shortest path, and establishing a reference rescue path set. According to the method, through dynamic analysis of building plan structure information, rooms and corridor intersections are abstracted as nodes, doors and channels are abstracted as edges with geometric weights, the accessibility of the edges is adjusted in real time after fire spreading or structure collapse occurs, impassable conditions are introduced during path calculation, and the path calculation accuracy is improved. And the real-time adaptability and rescue simulation precision of the path generation process are improved.

The invention discloses an intelligent household electrical appliance human engineering design optimization method and device based on a large model, and the method comprises the steps: collecting user motion and interaction data through a camera, a motion capture sensor and other devices, and extracting joint motion, operation frequency and other ergonomic features after time synchronization, coordinate calibration and noise cleaning; secondly, constructing a knowledge graph fusing anthropometry and biomechanics knowledge, performing graph embedding through graph convolutional network coding, and performing attention mechanism fusion with feature vectors; and then, training is carried out based on a large model of a Transform architecture in combination with ergonomic scoring constraints, and home appliance design parameters are optimized by using an evolutionary algorithm to generate a personalized scheme adapted to the height and operation habits of the user. According to the method, through cooperation of multi-modal data driving and domain knowledge, the comfort, accessibility and user adaptation precision of the intelligent household electrical appliance are remarkably improved.