1990 results about "State graph" patented technology

The invention provides a patient information collection and medical record construction system and method based on multiple rounds of dialogues, and the system comprises an intelligent guide interaction module which receives the natural language input of a patient; the context-aware question and answer engine adopts a dialogue state representation method based on a graph structure to construct entities, relationships and attributes of each round of dialogue into knowledge sub-graphs; the medical record information dynamic builder monitors an updating event of the dialogue state diagram in real time; the abnormal information detection module adopts a mixed conflict detection method combining rules and learning; and the medical knowledge graph support system maps the oral expression of the patient to a standard medical term system in real time through a multi-level semantic matching strategy. According to the invention, the doctor does not need to distract the record in the inquiry process, and can pay more attention to patient observation and clinical thinking. The quality of the medical record first draft automatically generated by the system is high, a doctor only needs to perform a small amount of auditing and supplementing, and the medical record writing time is greatly shortened.

The invention relates to the technical field of intelligent manufacturing and production management, in particular to a self-adaptive production scheduling system based on artificial intelligence, which comprises a data acquisition and reference construction module for analyzing process data to construct a directed acyclic graph representing a non-interference state as a reference map; the theoretical disturbance simulation module is used for converting the interference rule into a graph change instruction, generating a theoretical damaged state graph and obtaining a theoretical difference feature vector; the theoretical difference feature vector comprises, but is not limited to, a vector form obtained after a difference matrix is expanded according to rows or columns in terms of mathematical representation; the real deviation extraction module is used for collecting real-time state data to construct a real-time operation state diagram and calculating a real difference feature vector; a double-domain coupling decision module; an adaptive scheduling execution module; according to the method, the causal relationship is verified by comparing the form of theoretical deduction and actual observation, non-systematic noise is effectively filtered, and accurate response to real faults is realized while the stability of the production rhythm is maintained.

The invention discloses an intelligent question bank retrieval and recommendation system based on an artificial intelligence knowledge graph, and relates to the technical field of artificial intelligence education. Comprising a knowledge graph construction module which is used for processing original education data and constructing a neighborhood knowledge graph comprising a hard preposition relation and a soft incidence relation; the user knowledge state graph construction module is used for constructing a user personal knowledge state graph isomorphic to the domain knowledge graph, and dynamically calculating a mastery index of each knowledge node through a deep knowledge tracking model based on user historical answer data; according to the method, by constructing the domain knowledge graph containing the hard preposition relation and the soft incidence relation, discrete knowledge points are organized into the structured network conforming to the cognitive law, so that the system can understand and follow the internal logic between knowledge, and a learning path which is clear in organization, efficient and coherent is generated.

The invention discloses an industrial ERP dynamic scheduling optimization method based on deep reinforcement learning, and the method comprises the following steps: S1, obtaining to-be-scheduled industrial ERP data, and carrying out the preprocessing of the to-be-scheduled industrial ERP data; s2, constructing a heterogeneous graph structure, and setting node types and edge weight parameters in the graph; s3, performing multi-layer graph feature extraction on the initial scheduling state graph based on a graph attention mechanism to generate a node feature vector; s4, constructing a distributed reinforcement learning environment, and setting an independent strategy sub-network for each type of scheduling sub-tasks; s5, performing joint decision-making on the scheduling action to generate a scheduling scheme as a current round scheduling result; s6, calculating a reward value and feeding back the reward value to the corresponding strategy sub-network; and S7, repeatedly executing the steps S1 to S6, and generating a new scheduling action by using the updated strategy sub-network. According to the method, adaptive optimization of industrial ERP scheduling under multi-source heterogeneous data is realized, and the resource utilization rate and the order delivery efficiency are remarkably improved.

The invention provides a mountain power station current collection system state online monitoring method and device, and relates to the field of data processing. According to the method, distributed optical fiber data, environmental meteorological data and electrical measurement data are acquired to construct a multi-mode time sequence matrix, an operation state diagram is formed through graph structure representation and graph signal processing, a working condition mode cluster is obtained by combining dynamic clustering, a constraint diagram optimization framework is input, thermal, electrical and mechanical constraints are introduced, and an operation state track is generated. And constructing and fusing a non-stationary degradation path library and environment exposure history to form a dynamic degradation evolution model, and finally outputting in-service reliability and predicting residual life by a risk inference engine to realize online monitoring of the state of the current collection system of the mountain power station. By implementing the technical scheme provided by the invention, the accuracy of online monitoring of the state of the current collection system of the mountain power station is improved.

The invention discloses a research and development test bidirectional synchronization method based on artificial intelligence, and particularly relates to the technical field of artificial intelligence and software testing. Deep semantic modeling of research and development task behaviors is realized by constructing a multi-modal causal intention map, a test coverage blind area is inferred and recognized based on a graph neural network, a completion task is reversely generated, a complete state set is formed in combination with an original task set, and a condition control vector is adopted to drive generation of a structured test case. Collecting behavior and operation logs after the test is executed, detecting a mismatch area through modal alignment, generating a repair suggestion by utilizing a deviation correction model, and automatically writing back a task system; meanwhile, a state graph dynamic updating mechanism and a reverse task scheduling engine are introduced, and linkage feedback of test results on research and development tasks is achieved; the method can improve test coverage precision and research and development cooperation efficiency, and is suitable for agile development and continuous integration scenes.

Methods, systems, and computer readable media for rendering context-aware and interactive artificial intelligence-generated videos in real time. A talking face (“TF”) model may traverse from a first node to a second node of a state graph via an edge based on a TF instruction generated by an interaction model. The TF model may retrieve a transition video associated with the edge and a pre-computed video template associated with the second node from one or more TF databases. The pre-computed video template may include a plurality of masked video frames and a plurality of pre-computed mouth positions for each masked video frame. The TF model may inpaint a pre-computed mouth position into a masked region of each masked video frame to form a video frame stream. The interaction model may generate a video from the transition video and the video frame stream and present the video on a user device.

The invention discloses an API-oriented dynamic fuzz testing and vulnerability detection method, and belongs to the technical field of software security testing. The method comprises the following steps: extracting a dependency relationship, an input parameter, an output response and context state data of API calling, generating an initial API dependency graph, dynamically updating by capturing API state change in real time, forming an API state graph, and executing boundary-oriented variation based on parameter constraint characteristics, so as to obtain an API state graph; generating a variation test parameter, calling an API (Application Program Interface) of the variation test parameter to monitor a process memory behavior and response metadata, and generating a multi-dimensional abnormal signal; and performing mode matching on the abnormal signal and the vulnerability feature knowledge base, outputting a vulnerability type label and generating a path tracing report. According to the method, a context state sensing dynamic graph modeling technology is adopted, and a boundary-oriented intelligent variation strategy and multi-source abnormal behavior collaborative analysis are combined, so that precise vulnerability triggering, intelligent vulnerability judgment and complex scene coverage can be realized.

The invention relates to the technical field of government affair entrepreneurship guidance service intelligent dialogue systems, in particular to a dialogue intention recognition and correction method based on a knowledge graph, and the method comprises the steps: obtaining a voice signal and text information inputted by a user, and carrying out the cross-modal fusion processing to generate a semantic vector; carrying out evolution modeling on the dialogue state by adopting a graph neural network, and outputting a dynamic dialogue state graph containing a time sequence relationship between a user entity and a business object; taking the semantic vector and the focus information as query conditions, executing dynamic subgraph sampling from the government affair knowledge graph, and generating an intention correction strategy vector; and finally, utilizing reinforcement learning to generate personalized guide statements and outputting the guide statements to the user. According to the method, through deep coupling of semantic understanding, state tracking, knowledge reasoning and decision optimization, real-time perception and active completion of dynamic updating of hidden nodes and policies of the government affair process are achieved, and the intention recognition accuracy and systematic adaptive capacity are effectively improved.

The invention relates to the technical field of cognitive behavior therapy, in particular to a scheme intelligent pushing and task self-adaptive management method based on cognitive behavior therapy, which comprises the following steps: S1, acquiring physiological signals, behavior logs and environment interaction data of a user through a multi-modal data acquisition module; s2, constructing a dynamic psychological assessment model based on a cognitive behavior theory, and fusing multi-source data by adopting a Bayesian network to generate a user cognitive state map; s3, generating a personalized intervention scheme through a reinforcement learning algorithm according to the cognitive state map and a preset CBT intervention rule base; according to the method, through multi-modal data acquisition, dynamic psychological assessment, reinforcement learning scheme generation, task decomposition optimization, difficulty adaptive adjustment and digital twinborn simulation, full-process closed-loop management from data acquisition to intervention optimization is constructed, and efficient, safe and personalized cognitive behavior treatment scheme intelligent pushing and task adaptive management are realized.

The invention discloses a design manuscript code generation and visual error iterative correction method and system based on multi-agent cooperation, and aims to solve the problems of low front-end code generation efficiency, poor visual restoration degree and lack of an automatic correction mechanism in the prior art. The system constructs a four-node state diagram based on LangGraph, automatically initializes a series engineering scaffold, obtains design data and generates a structured protocol, and generates a component code based on the protocol and performs visual verification and iterative correction on four processing nodes. The method comprises the following steps: dynamically generating an engineering scaffold through a large language model agent, converting Figma design data into a structured protocol, and carrying out iterative optimization by utilizing a judger-repairer mode until an error is lower than a threshold value; according to the method, high-fidelity automatic delivery from the design manuscript to the code is realized, and the executable front-end engineering code and the visual reduction degree analysis report are output.

The invention discloses a space chain reasoning-based space intelligent multi-modal space understanding and planning method, relates to the field of space intelligence, solves the problem that the existing space intelligent multi-modal reliability and interpretability are insufficient, and comprises the following steps: S1, obtaining multi-modal space data; s2, constructing a spatial semantic state diagram; s3, task nodes are extracted, the spatial semantic state diagram is traversed, a logical reasoning list is constructed, and a reasoning track is obtained; s4, performing simulation operation on the motion subject model in the space world model based on the reasoning track, recording a simulation position and a simulation operation state, and optimizing the reasoning track to obtain an implementation track; s5, constructing a space cost function, and dynamically optimizing the implementation trajectory according to the space cost function; s6, the reasoning track is corrected; according to the method, the reliability and interpretability of space intelligence are effectively improved in a space chain reasoning mode.

The invention discloses an automatic compliance examination method and system based on graph retrieval enhancement. The method comprises the following steps of: obtaining a billing specification and a historical inquiry report as input documents; the input document is subjected to dual-mode document segmentation based on a large language model, logic blocks and physical blocks are generated, and each logic block is a continuous page unit and is attached with a content abstract generated by the model; constructing a precedent graph, extracting key legal elements in the historical inquiry report through multi-stage recursion, generating semantic network nodes carrying traceability identifiers, and establishing cross-document semantic association; and constructing a state graph, automatically identifying chapters and logic and semantic relationships among the chapters based on a document hierarchical structure, and generating a machine-readable structured index and the like. According to the method, the long text semantic understanding depth, the cross-section consistency and the legal reasoning accuracy are remarkably improved, the manual review cost is reduced, and the method is suitable for a listing compliance review scene under a registration system.

The invention relates to the technical field of water conservancy project intellectualization, and discloses a water affair system digital twin-driven intelligent early warning and self-adaptive regulation and control platform, which comprises the following steps of: acquiring multi-source data related to operation of a water affair system through Internet of Things equipment, and generating a data set in a uniform format; integrating the multi-source data through a cross-domain data fusion engine to generate a fusion feature set; analyzing a causal relationship among variables in the multi-source data through a time sequence causal inference engine, and generating a dynamic causal graph; a system operation mode is optimized through information interaction between nodes, and distributed cognitive data is generated; generating a state atlas of the water affair system based on the topological network and the dynamic causal graph through a synchronous state atlas visualization engine; and generating a regulation and control instruction according to the multi-target collaborative regulation and control strategy, and updating the data set. According to the method, the abnormal events in the water affair system can be accurately identified through the cross-domain data fusion engine and the time sequence causal inference engine.

The invention relates to the technical field of short message transmission optimization, and discloses an intelligent short message resending system and method based on receipt state recognition. A communication data acquisition module of the system obtains historical short message transmission records in a preset time period of a target communication area and generates a communication state data set; a feature extraction storage module performs receipt feature extraction on the data set to obtain a historical receipt feature set and establish a receipt state change matrix; the topology modeling module constructs a terminal state atlas according to the matrix, and calculates the topology aggregation degree of the historical short message failure events to determine an abnormal characteristic index set; a real-time analysis module extracts real-time receipt features of the current short message transmission data to generate real-time state vectors, and calculates a topology correlation degree with the abnormal feature index set in a terminal state graph to generate real-time risk factors; and the decision generation module generates a short message resending strategy instruction according to the real-time risk factor, the historical receipt feature set and the receipt state change matrix.

The invention discloses a civil engineering construction task automatic scheduling method and system. The method comprises the steps that a heterogeneous graph model representing construction task elements and the incidence relation of the construction task elements is constructed; processing the heterogeneous graph model by using a graph neural network to mine a dependency relationship between processes and extract risk association features related to uncertainty to form a construction state graph; constructing a Bayesian network based on the construction state atlas; probabilistic reasoning is carried out through a Bayesian network, the influence of uncertain factors is quantified, and key information representing risk conduction is fed back to the graph neural network so as to dynamically correct recognition of the graph neural network on a process dependency relationship; the dependency constraint output by the graph neural network and the risk quantification result output by the Bayesian network are fused, and a construction task scheduling scheme is generated; and synchronously updating the heterogeneous graph model, the graph neural network and the Bayesian network according to field construction feedback data to realize closed-loop iterative optimization of a scheduling scheme.

The invention provides an interactive command line tool-oriented intelligent task planning and execution method and system, and the method comprises the steps: S1, generating a corresponding command sequence template for each operation of a target tool through a first large language model, and constructing an executable knowledge base; s2, calculating the semantic similarity between the task description and each command sequence template to obtain a candidate set, and selecting the most relevant template as a retrieval context by using a second large language model; s3, modeling a task execution process as a state diagram search problem, selecting candidate nodes from reachable nodes of a current node based on an algorithm, and selecting an optimal node according to a current state and a retrieval context by utilizing a second large language model; s4, executing an execution action corresponding to the optimal node, and analyzing tool output by using a second large language model to adjust a heuristic value of the corresponding node and execute a corresponding rollback strategy; and S5, repeating the steps S3 to S4 until the reasoning result shows that the task reaches the target node or the number of attempts exceeds a preset threshold.

The invention relates to the technical field of psychological health intelligent evaluation and collaborative intervention, and discloses a family-school psychological health collaborative method and system.The method comprises the steps that a multi-dimensional data model is built based on the psychological theory, and family-school double-end data collection is achieved through standardized indexes; constructing a multi-dimensional psychological state diagram structure, and realizing home-school data deep fusion by adopting a time sequence dynamic coding method; a time sequence modeling and multi-level risk assessment method is adopted, and a dynamic mental health risk assessment result and an intelligent early warning signal are output; a cross-platform collaborative intervention management method is adopted, and a personalized intervention strategy and a collaborative task allocation scheme are output; and a multi-dimensional effect evaluation and reinforcement learning optimization method is adopted to realize continuous optimization of an intervention strategy. Through artificial intelligence technologies such as the dynamic graph neural network and deep reinforcement learning, intelligent fusion and cooperative intervention of family-school psychological health data are realized, and the early recognition accuracy and intervention effect of psychological health problems are improved.

1. The name of the design product: custom window of graphical user interface for customizing display layout of electronic device. 2. The use of the design product: electronic device. 3. The design points of the design product: in graphical user interface. 4. The picture or photo that best indicates the design points: design 1 front view. 5. Design 1 is designated as the basic design. 6. The use of the graphical user interface: enables the user to customize the display layout of data received from one or more inputs connected to the medical console. 7. The human-computer interaction mode of the graphical user interface: interact through the touch screen of the medical console, when the number of screens in the upper right corner of the graphical user interface is selected as "1", the display layout shown in each design front view appears; when the number of screens in the upper right corner of the graphical user interface is selected as "2", the display layout shown in each design change state diagram 1 appears; when the number of screens in the upper right corner of the graphical user interface is selected as "3", the display layout shown in each design change state diagram 2 appears; when the number of screens in the upper right corner of the graphical user interface is selected as "4", the display layout shown in each design change state diagram 3 appears. 8. Other circumstances that need to be explained: the dashed line does not constitute part of the claimed design.

The invention provides an urban road network traffic congestion state prediction method and system based on deep learning, and relates to the technical field of intelligent traffic control, and the method comprises the steps: obtaining congestion state image data of an urban road network, and carrying out the preprocessing to extract structured road congestion information; constructing a road network topological graph based on the road congestion information, and performing structure sensing state coding on the road network topological graph to generate a structure sensing state vector fusing local congestion features and global topological position information; and inputting the structure perception state vector into an improved deep learning prediction model, performing feature deepening and enhancement through a graph neural network which dynamically adjusts a graph structure and aggregates node information, and finally outputting road network congestion state prediction results of a plurality of time scales in the future. According to the method, accurate prediction of the urban road network congestion state can be realized, and reliable technical support is provided for intelligent traffic management, resource scheduling optimization, travel path planning and the like.

The invention relates to the technical field of electric power data, and discloses a multi-source electric power appeal data classification management method, system and device and a medium, and the method comprises the steps: constructing a multilayer graph structure according to an obtained multi-source electric power appeal data set; according to the multi-layer graph structure, a search sample set in which each search sample comprises a color vector, a state vector and a fitness state is constructed, the color vectors are determined by semantic features, and the fitness states are set to reflect the semantic connectivity of a semantic graph, the state fluctuation amplitude of a state graph and the adjacency density of a spatial adjacency graph; an improved peacock optimization algorithm is adopted to conduct iterative updating on the search sample set, the optimal fitness state of each search sample is obtained so as to determine classification labels corresponding to the search samples, and the improved peacock optimization algorithm is set to conduct iterative updating on each search sample according to the disturbance guide vector and the gravitational gradient direction. According to the method provided by the invention, the accuracy and adaptability of power demand sorting are improved.

The invention relates to an intelligent state monitoring method and system for a high-voltage isolation switch operating mechanism, and relates to the technical field of power equipment fault diagnosis, and the method comprises the steps: enabling a collected real-time sequence to update a standard operation response potential energy surface, obtaining a real-time operation response potential energy surface, generating a phase response difference tensor through data projection, and obtaining a real-time operation response potential energy surface; and depicting deviation characteristics of actual and standard operation states. And then, in the constructed graph attention network, on the basis of a real-time sequence, monitoring parameters and video motion characteristics, constructing a multi-modal sequence state graph, and determining a state typing result through fine-grained recognition and interpretable analysis of a fault. And finally, introducing a cyclic fingerprint residual model, performing clustering optimization based on a real-time sequence, determining an abnormal deviation index, and combining the abnormal deviation index with a state typing result to determine a fault diagnosis result. Therefore, the accuracy of intelligent state monitoring of the high-voltage isolation switch operating mechanism can be effectively improved, and the method can be suitable for non-ideal environments, low dominant anomalies and other conditions.

The invention discloses a road maintenance construction equipment operation safety evaluation method and system, and relates to the technical field of equipment safety evaluation, and the method comprises the steps: collecting equipment multi-source operation state data, and carrying out the normalization processing, and forming a standardized working condition vector sequence; by taking each time sequence vector in the working condition vector sequence as a graph node, constructing a weighted adjacency matrix to generate a working condition evolution state graph; constructing a task feature vector according to the demand of a to-be-executed task, matching atlas nodes, calculating a path comprehensive score, and screening an optimal evolution path; all nodes in the optimal evolution path are mapped to a construction area, and the comprehensive score of the path is corrected by fusing environmental factors; and grading and classifying the corrected score and outputting a safety level and a control strategy. According to the invention, through dynamic risk correction and evolution path pre-judgment, closed-loop intelligent decision-making of construction safety is realized, and early warning timeliness and environmental adaptability are improved.

The invention discloses a chip circuit state self-diagnosis method based on a graph neural network, and the method comprises the following steps: obtaining a chip netlist, layout and process information, collecting multi-source state data to construct a multi-domain coupling graph, and generating a multi-scale calculation tree based on a critical path; standardizing key node noise data, extracting noise fingerprints, constructing a noise consistency graph, and aligning the noise consistency graph with the multi-domain coupling graph to form node features; inputting the multi-scale calculation tree into an improved GFT model, and generating unified graph representation through structure token coding and feature fusion; constructing a multi-scale transition graph neural network, simulating multi-domain state propagation and outputting an abnormal score; anti-fact operation and reversible disturbance are applied to the high abnormal score nodes to generate an anti-fact state diagram, and a fault causal chain is constructed; and performing topological variation inversion based on the abnormal score and the causal chain to obtain a fault position, a fault type and a fault degradation trend. According to the invention, high-precision explainable chip fault self-diagnosis can be realized.

The invention discloses a surface defect detection system based on a microcrystalline glass panel finished product, and relates to the technical field of surface defect detection. The defect full-dimensional dynamic state map construction module is used for constructing and updating a defect full-dimensional dynamic state map containing defect three-dimensional coordinate information, multi-modal feature confidence information, virtual-entity parameter deviation information and feature definition-registration precision linkage threshold information in real time; the structured light three-dimensional contour acquisition module is used for acquiring three-dimensional contour data of the surface of the microcrystalline glass panel; the multi-modal feature fusion module is used for collecting two-dimensional image data of multi-light-source imaging and extracting multi-modal feature information; the digital twin virtual debugging module is used for constructing a virtual detection environment based on the defect full-dimension dynamic state graph; the problem of feature confusion caused by light source combinational logic solidification in an existing multi-light-source switching system is effectively solved.

The invention discloses an intelligent customer service question and answer optimization method and system based on large model driving, and relates to the technical field of data processing.The method comprises the steps that a semantic flow self-supervision state machine is used for generating a semantic attribution graph and extracting core semantic factors, reverse semantic generation and cross-factor compensation are executed, and a structured target semantic set is constructed; a multi-branch task state diagram is constructed through behavior path inversion, and after a main task state is recognized, two-dimensional semantic projection is carried out, and a strategy tensor is generated; generating a polymorphic answer and constructing a reply text based on the strategy tensor, the main task state and the semantic set; and calculating a deviation vector through a user feedback path and updating the large model. The technical problem that an existing intelligent customer service system is low in semantic understanding precision, single in answer and lack of individuation, and consequently the question and answer response speed and precision are low is solved, and the technical effect that the semantic understanding precision and the answer diversity are improved through a dynamic optimization mechanism driven by a large model, so that the question and answer response speed and precision are improved is achieved.

One or more embodiments of the invention provide a task processing method. The method comprises the following steps: receiving a to-be-executed task; the to-be-executed task is one of tasks which need to be executed and are determined by analyzing a question input by a user; creating a target node corresponding to the to-be-executed task in a state graph; the state atlas corresponds to the question and comprises a plurality of nodes, and edges among the nodes are used for representing dependency relationships among tasks corresponding to the nodes; determining a target tool or a target agent according to the to-be-executed task, and executing the to-be-executed task by using the target tool or the target agent; and under the condition that the execution of the to-be-executed task fails, recovering the to-be-executed task according to the state graph.

Simulation and validation of neural network systems is provided. In various embodiments, a description of an artificial neural network is read. A directed graph is constructed comprising a plurality of edges and a plurality of nodes, each of the plurality of edges corresponding to a queue and each of the plurality of nodes corresponding to a computing function of the neural network system. A graph state is updated over a plurality of time steps according to the description of the neural network, the graph state being defined by the contents of each of the plurality of queues. Each of a plurality of assertions is tested at each of the plurality of time steps, each of the plurality of assertions being a function of a subset of the graph state. Invalidity of the neural network system is indicated for each violation of one of the plurality of assertions.

The invention belongs to the technical field of battery changing cabinets, and discloses a multi-modal data fusion intelligent battery changing cabinet battery state monitoring system and method, and the method comprises the steps: collecting the electrical parameters, environmental parameters and use parameters of a battery; performing sliding window monitoring based on the electrical parameters, and generating a working condition time slice; performing multi-modal feature extraction on the working condition time slice to generate a node attribute vector; node attribute vectors of the latest N working condition time slices are extracted, and a battery state diagram is generated; performing feature learning on the battery state diagram by using a diagram neural network to generate a diagram embedding vector; and inputting the graph embedded vector into the health prediction model to obtain a battery health state value, and generating graded early warning information and maintenance suggestions. The battery state evolution characteristics are effectively captured through dynamic working condition segmentation and graph neural network time sequence modeling, preventive maintenance is achieved, and the problems that a traditional method is single in dimension and lacks time sequence characteristic consideration are solved.

The invention discloses a heterogeneous hardware computing power scheduling method and device, a program product and a medium, and relates to the technical field of computing power scheduling. The method comprises the following steps: collecting a current instruction execution stream and a current temperature of each heterogeneous hardware unit; predicting an expected load mode of each heterogeneous hardware unit based on the current instruction execution stream; determining a thermal expected state of each heterogeneous hardware unit based on the expected load mode and the current temperature; based on the thermal expectation state of each heterogeneous hardware unit, constructing a global thermal expectation state map; when a to-be-scheduled target task is received, determining a first hardware unit in each heterogeneous hardware unit based on the target task and the global thermal expected state graph; the target task is scheduled to the first hardware unit to be executed, and the task queue priority of at least one second hardware unit physically adjacent to the first hardware unit is adjusted according to the global hot expected state graph. The method has the effect of improving the accuracy of computing power scheduling.