47 results about "Probabilistic graph" patented technology

The invention discloses an SVG valve hall cooling efficiency evaluation method and system based on a probabilistic graph model, and the method comprises the steps: obtaining original time sequence data which is obtained through the collection of a cooling system multi-parameter monitoring sensor group disposed in an SVG valve hall in continuous T sampling periods; preprocessing the original time series data to obtain a credible time series data set; constructing a Bayesian network topological structure comprising three-level nodes of an environment layer, a component layer and an efficiency layer and causal dependence edges, and optimizing parameters of the Bayesian network topological structure by adopting a maximum likelihood estimation method to form a dynamic Bayesian network model after parameter calibration; and the credible time sequence data set is used as an evidence variable to be input into the Bayesian network model after parameter calibration, calculation is carried out through a belief propagation reasoning algorithm, a final control instruction set is generated through probability weighted scoring processing, the final control instruction set is fed back to a valve group monitoring system, and early warning and automatic load reduction are achieved. The problems of large evaluation deviation and early warning lag in the prior art are solved.

Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support ontology driven processes to create digital twins that extend the capabilities of knowledge graphs. A dataset including an ontology and domain data corresponding to a domain associated with the ontology is obtained. A knowledge graph is constructed based on the ontology and the domain data is incorporated into the knowledge graph. The knowledge graph is exploited to derive random variables of a probabilistic graph model. The random variables may be associated with probability distributions, which may include unknown parameters. A learning process is executed to learn the unknown parameters and obtain a joint distribution of the probabilistic graph model, which may enable querying of the probabilistic graph model in a probabilistic and deterministic manner.

The invention discloses a super capacitor-lithium battery hybrid energy storage life optimization method based on variational Bayesian inference, and relates to the technical field of battery energy storage. The method comprises the following steps: step 1, acquiring operation data of the super capacitor-lithium battery hybrid energy storage system under rated power charge-discharge cycle; a multi-scale decommissioning-state initialization tensor is obtained; step 2, defining a hidden variable set, constructing a three-branch hierarchical directed acyclic probability graph model, and adopting variational Bayesian inference to output approximate posteriori distribution and variational lower bound of the hidden variable set; 3, calculating expectation and variance of approximate posteriori distribution of the hidden variable set; and dividing the real-time power demand into a super-capacitor power instruction and a lithium battery power instruction according to the power distribution coefficient, and issuing the super-capacitor power instruction and the lithium battery power instruction to super-capacitor-lithium battery hybrid energy storage in real time. The lithium battery decline can be effectively delayed, and the service life and reliability of the system are improved.

The invention discloses a dynamic monitoring and accounting system for organic carbon reserves of lake and reservoir sediments, belongs to the technical field of carbon cycle monitoring, and aims to solve the problems of spatial-temporal resolution contradiction and insufficient multi-source data fusion caused by a single data source in a traditional method. A space-time continuous fusion data product is generated in combination with an adaptive fusion algorithm, the system comprises a multi-source data acquisition module, a preprocessing module, an uncertainty quantification module, a fusion processing module and a verification feedback module, deep fusion of space-based remote sensing, foundation in-situ and experimental analysis data is achieved, and through independent actual measurement data verification and parameter feedback optimization, the space-time continuous fusion data product is obtained. And the system forms a precision improvement closed loop, so that the stability and accuracy of long-term operation are ensured.

The application relates to the field of signal processing and discloses a unified probability graph computing architecture power consumption optimization method, system, medium and equipment, which comprises the following steps: a unified probability graph for multiple tasks is used to perform sparse optimization on message passing data flow by dynamically regulating the message passing data flow in a probability iteration calculation process; and based on the optimized message passing data flow, the activation state of a memory-computing integrated array node is dynamically selected by a computing circuit of a memory-computing integrated characteristic device to improve the adaptability of the computing architecture and hardware resources. The application reduces the calculation-storage scale by deleting probability messages with low reasoning contribution and dynamically regulating the message passing data flow in the probability iteration calculation process, and the adaptability of the computing architecture and hardware resources is improved by dynamically selecting the activation state of the memory-computing integrated array node, so that the power consumption is further reduced.

The invention relates to a channel estimation method and system based on a storage and calculation integrated device, processing equipment and a storage medium, and the method comprises the steps: obtaining a pilot frequency structure matrix and a probability graph model of a to-be-measured channel estimation task, and obtaining a channel estimation result according to a pilot frequency signal observed by a receiving end and a variance sum of channel uniform quantization; determining a state value initial probability of each variable node in the probabilistic graph model; constructing a node state matrix corresponding to each undirected edge in the probability graph model; constructing a driving vector corresponding to each undirected edge in the probability graph model; based on the node state matrix corresponding to each undirected edge, storage and calculation all-in-one device configuration is carried out, and a storage and calculation all-in-one device sub-array corresponding to each undirected edge is obtained; and determining a continuous channel estimation result in the to-be-measured channel estimation task according to the storage and calculation all-in-one device sub-array, the driving vector and the confidence coefficient of message transmission between each variable node and each check node connected through the undirected edge, and the method can be widely applied to the technical field of signal processing.

The invention provides a heterogeneous storage system, a data synchronization method and device, electronic equipment and a medium, and relates to the technical field of artificial intelligence and the like. According to the specific implementation scheme, a memory object layer comprises a probability graph model and a dirty node subset, the probability graph model comprises a node set and an edge set, and the dirty node subset is used for recording node identifiers of modified nodes or edges in the probability graph model; the persistent storage layer comprises a vector database and a graph database which are physically separated; the data mapping module is used for mapping the feature vectors and the uncertainty parameters to a vector database and mapping connection weights and causal attributes to a graph database; the synchronous controller is configured to monitor a space index and a time index in real time, the space index is determined based on the current state of the dirty node subset, and when it is detected that the space index or the time index meets a preset batch submission condition, the changed data of the memory object layer is synchronized to the persistent storage layer based on the dirty node subset.

The application relates to the technical field of automation operation and maintenance, and discloses an intention recognition method and system based on dynamic ontology evolution and multi-agent, which comprises the following steps: receiving a natural language instruction, performing entity extraction and probabilistic linking by using a dynamic ontology knowledge base, and generating an initial intention based on predicate analysis; automatically completing missing key slots by using a probabilistic graph model, and generating a standardized intention; decomposing the standardized intention into an atomic subtask sequence, dynamically matching an execution agent based on an agent capability-demand matrix, generating a collaborative workflow, controlling the execution agent to call an atomic tool to execute a task, performing causal correlation analysis on multi-source results according to logical relations between ontology instances, generating a structured reasoning chain, and feeding back; and extracting a new treatment script based on execution feedback by using an evolution engine, and updating an ontology knowledge base and an agent confidence degree. The application can realize accurate understanding, automatic execution and adaptive evolution of a knowledge base of a fuzzy operation and maintenance intention.

The invention provides an automatic metadata recognition and extraction method for remote sensing images and spatial data, and belongs to the technical field of remote sensing measurement. Multi-source remote sensing image data are collected and preprocessed to establish a multi-modal image set under a unified geographic coordinate system, and an image pyramid is constructed by using a GPU accelerated block parallel processing architecture; a topographic feature extraction algorithm based on topology durability is used to identify stable topographic markers, a geometric registration feature extraction and matching network is constructed, a structured prediction framework based on a probability graph model is used to carry out accurate registration, and adaptive filtering and multi-modal data fusion are carried out on a registered image set. A semi-supervised classification algorithm based on random walk is used for carrying out ground feature classification, metadata elements are automatically identified according to a ground feature classification result and stable topographic markers, and the technical problem that the registration precision is insufficient due to the fact that topographic feature extraction is prone to noise interference in multi-mode remote sensing image registration is solved.

Embodiments of the present application relate to the field of photovoltaic technology, and disclose a photovoltaic monitoring system abnormal test data generation method, device, equipment and medium, the method comprises the following steps: constructing a GAN model, constructing a fusion framework of a knowledge graph and a probabilistic graph model, including constructing an ontology model in the field of photovoltaic, and establishing a probabilistic graph model according to the constraint relationship in the ontology model, iteratively training the GAN model to convergence, and then outputting a preset number of evaluation test samples of a predetermined photovoltaic fault type, inputting the preset number of evaluation test samples into the probabilistic graph model, and outputting a multidimensional influence probability corresponding to each evaluation test sample, obtaining a business impact degree corresponding to the evaluation test sample according to the multidimensional influence probability, and obtaining test data by screening the evaluation test sample according to the business impact degree. The test data generated by the embodiments of the present application can be widely applied to the test and verification of data cleaning, fault tolerance processing, fault detection and other functions of the photovoltaic monitoring system.

The application relates to the technical field of relationship graph data clustering, and provides a relationship graph data clustering method and system based on a non-parametric Bayesian method, which comprises the following steps: obtaining relationship graph data; initializing model parameters according to a probabilistic graph model; using multiple sampling algorithms to iteratively sample samples of the model parameters based on the relationship graph data; selecting a cluster division sample according to a maximum likelihood function mode based on the model parameters obtained through sampling, so as to obtain a clustering result; and the prior probability of the cluster division sample is a non-parametric Bayesian prior. The number of clusters can be automatically inferred in the parameter inference process, and manual setting is not needed.

The present application relates to the technical field of artificial intelligence monitoring, and more particularly to a disease early warning method and system fusing a deep probabilistic graph model and Bayesian inference, which inputs multi-source time series monitoring data and static attribute features into a deep generative model, maps to an independent latent space and decouples output pathological, environmental and noise variables, purifies the pathological variables using the latter two, acquires and propagates structure particles based on the purified features, inputs a continuous time evolution model to generate multiple epidemic evolution trajectories, and finally calculates a comprehensive risk value through a risk sensitive evaluation function to trigger an early warning, thereby achieving accurate separation of environmental drift and random noise from mixed signals and significantly reducing the false positive rate. Meanwhile, multiple propagation hypotheses are deduced in parallel, and the weight of high-risk trajectories is amplified through nonlinear aggregation, thereby ensuring that long-tail disaster risks can be effectively captured in the face of uncertainty.

The invention relates to the technical field of network security, and discloses a social simulator cognitive intervention method and system, and the method comprises the steps: generating a group topology based on the user interaction data of a real social network; based on the initial evaluation signal and a Bayesian optimization method, dynamically adjusting the configuration proportion of the agent node; reasoning the marginal probability of intrusive information of each agent node in the group topology based on a probabilistic graph model, screening candidate intervention nodes according to the marginal probability, deploying intervention type agent nodes based on the adjusted configuration proportion, and performing intervention on the candidate intervention nodes; the intervention result is evaluated to generate an evaluation signal, and the evaluation signal is fed back to the dynamic adjustment step to replace the initial evaluation signal. According to the method, intervention subjects of different roles can be effectively organized based on large-scale data of a complex network environment, the putting proportion is dynamically optimized, and simulation prediction is carried out on the to-be-intervened nodes based on probability graph reasoning.

The application provides a high-order interaction prediction method and device with mixed graph deep learning, which first constructs a drug molecule graph, a microorganism weighted graph, a disease weighted graph and a supergraph connecting the three based on multi-source heterogeneous data such as drug molecular structure, microorganism classification information and disease semantic network, forming a mixed graph structure. Subsequently, through a mixed graph deep learning module fusing a graph convolution network and a supergraph neural network, nonlinear structure features and high-order interaction features of each entity are extracted, and the adaptive fusion of the features is realized by using an attention mechanism. Then, the fused deep features are mapped to the prior expectation of the latent factor matrix in the Bayesian logic tensor decomposition model, a probabilistic graph model is constructed, and the joint adaptive inference of the model parameters, latent variables and deep learning mapping is carried out through a variational expectation maximization algorithm, so that the high-order correlation probability prediction of the whole tensor space is realized without negative sampling.

The invention discloses an analog circuit implementation method and device based on a unified probability graph computing architecture, equipment and a medium, and the method comprises the steps: building a unified signal model based on the common characteristics of a baseband signal processing task; representing the unified signal model as a unified probability graph model; obtaining an initialization probability of a variable node state value in the unified probability graph model; a storage and calculation integrated characteristic device is set, the storage and calculation integrated characteristic device comprises an initial input array, an iterative calculation array and a result output array, and node state values of the initial input array, the iterative calculation array and the result output array are configured based on the unified probability graph model; a driving vector is obtained based on the initialization probability to serve as input of the initial input array, the iterative calculation array is used for completing iterative calculation of mutual message passing of variable nodes and check nodes in the unified probability graph model, and when an iteration stop condition is met, the result output array outputs a processing result to complete a baseband signal processing task.

The invention discloses an urban bridge and tunnel group vehicle load modeling method and system based on a depth probability graph model, and belongs to the technical field of urban bridge and tunnel group health monitoring. The method comprises the following steps of: 1, preprocessing acquired historical data of a dynamic weighing system and acquired data of each passing vehicle; 2, time-space factors are collected; step 3, establishing a bridge-tunnel network based on the preprocessed data in the step 1 and the space-time factors in the step 2; 4, constructing a vehicle load depth probability model based on the bridge-tunnel network established in the step 3; 5, training based on the vehicle load depth probability model in the step 4; step 6, generating a load of a random traffic flow vehicle based on the model trained in the step 5; and 7, carrying out random traffic flow vehicle load deduction based on the model trained in the step 5. The method is used for solving the problems of bridge and tunnel operation and maintenance and bridge and tunnel reliability analysis, the influence of time difference on the model, and unclear correlation among nodes in the urban bridge and tunnel network.

This application relates to the field of geological engineering technology. To address the problems of time-consuming parameter learning and errors caused by reliance on subjective experience in undirected probabilistic graphical geological model construction, a method for constructing undirected probabilistic graphical geological models based on spatial location encoding is disclosed. This method includes: determining the geological modeling range based on known borehole data; discretizing the geological modeling range into a grid structure; filling the corresponding grid with stratigraphic data from the known borehole data according to soil type; spatially encoding the unknown grids based on the spatial relationship between the grid structure and the known borehole data; assigning an initial state to the unknown grids based on the spatial relationship between the grid structure and the known borehole data to obtain the initial stratigraphy and constructing an undirected probabilistic graphical geological model; and constructing a data-driven parameter optimization method based on the spatial location encoding results to learn the parameters of the undirected probabilistic graphical geological model and obtain the optimal parameters. This method improves the accuracy of undirected probabilistic graphical geological model construction.

The invention relates to the technical field of Internet of Things and machine learning, and particularly discloses a laboratory safety whole-process dynamic management system based on the Internet of Things. The system comprises an Internet of Things sensing layer, an edge calculation layer, a data fusion and knowledge construction layer, a dynamic risk assessment and decision-making layer and an execution and feedback layer. A dynamic security knowledge graph is constructed by fusing multi-source heterogeneous data, risk assessment is performed by integrating mode recognition and probability graph reasoning, a hierarchical management and control instruction is generated, and finally closed-loop management is formed through an execution layer, so that real-time and comprehensive perception, accurate risk assessment and adaptive dynamic management and control of a laboratory security state are realized.

The application belongs to the technical field of cloud computing operation and maintenance, and particularly relates to an alarm intelligent root cause analysis and self-recovery method for guaranteeing the continuity of a legal service platform, comprising the following steps: receiving heterogeneous alarms and performing standardized processing to generate a standard event model; based on feature fields in the standard event model, matching is performed in a preset routing rule library to determine a target analysis module; multi-dimensional diagnostic data associated with the standard event model is collected through the target analysis module, root cause reasoning is performed using a probabilistic graph model, a diagnostic conclusion with a confidence score and a candidate self-recovery strategy are generated; pre-checking is performed according to the risk level of the candidate self-recovery strategy and a security strategy library, if the checking passes, an execution instruction is generated, and if the checking fails, an artificial review process is triggered; the execution instruction is executed and the result is fed back to a knowledge base optimization model and a rule. The application can solve the problems of slow alarm response, difficult experience sedimentation and high operation risk.

The invention provides an on-line monitoring method for abnormal vibration of thermal energy storage equipment, and belongs to the technical field of thermal energy storage equipment. A multi-source sensor array comprising low-frequency and high-frequency vibration sensors is arranged at key parts of the thermal energy storage equipment, and noise interference is suppressed in a differential arrangement mode; a multi-source coupling vibration signal is separated by using an independent component analysis algorithm, time-frequency analysis is performed by combining wavelet transform to extract vibration characteristics, and a phase change vibration characteristic mechanism equation is established to calculate material physical parameter changes in a phase change process. A neural network identification model based on dynamic topology reconstruction sparse connection learning and probability graph model structured prediction is constructed, and a vibration anomaly discrimination threshold system is established to realize anomaly early warning and adaptive optimization. The technical problem that the abnormal vibration mode is difficult to accurately separate and identify by the multi-source coupling vibration signal in the phase change process of the thermal energy storage equipment is solved.

This invention discloses a hyperspectral image classification method and device based on a subgraph-dependent neural network. The method includes: inputting hyperspectral image data; optimizing the matrix structure using a spatial-spectral unified adaptive probabilistic graph convolutional network to construct an overall graph structure; dividing the overall graph structure into subgraphs, identifying optimal probabilistic connection matrices and corresponding feature matrices for adjacent subgraphs; embedding degree information into the feature matrices as residuals to construct residual features; constructing a set of clustering performance metrics and dynamically comparing them with a preset adaptive adjustment feedback threshold to obtain the optimal graph convolutional layer; generating a low-dimensional absolute position sequence through absolute position encoding, concatenating it with the node feature vectors output from the subgraph convolution as input to a Transformer encoder, fusing structural and spatial position information, and finally outputting the classification result. This invention can fully exploit the spatial, spectral, and structural features of hyperspectral images, significantly improving the accuracy and robustness of classification.

The application discloses a system fault positioning method and device based on a Bayesian network, relates to the technical field of fault detection, and comprises the following steps: a fault tree and a Bayesian network are adopted to construct a fault tree for an operating system and are converted into a Bayesian network, so that a fault event of the operating system is represented by a leaf node in the Bayesian network, and a fault reason causing the fault event is represented by a root node in the Bayesian network; since the Bayesian network belongs to a kind of probabilistic graph model, a fault probability of the root node in the Bayesian network can be quantified; when a target fault event occurs in the operating system, a target fault probability of a leaf node corresponding to the target fault event is determined based on a prior fault probability of the root node in the Bayesian network, and a posterior fault probability of the root node is back calculated, so that the posterior fault probability of the root node is used to accurately locate the final fault reason causing the target fault event, and the accuracy of system fault reason positioning is improved.

The application discloses a simulation circuit implementation method and device based on a unified probability graph calculation architecture, equipment and a medium, comprising: establishing a unified signal model based on the common characteristics of baseband signal processing tasks; the unified signal model is characterized as a unified probability graph model; the initialization probability of the variable node state value in the unified probability graph model is obtained; a storage and calculation integrated characteristic device is set, the storage and calculation integrated characteristic device comprises an initial input array, an iterative calculation array and a result output array, wherein the node state values of the initial input array, the iterative calculation array and the result output array are configured based on the unified probability graph model; the driving vector is obtained as the input of the initial input array based on the initialization probability, the iterative calculation array is used to complete the iterative calculation of the mutual message transmission between the variable nodes and the check nodes in the unified probability graph model, when the iteration stopping condition is met, the result output array outputs the processing result, and the baseband signal processing task is completed.

The invention relates to a channel estimation simulation circuit implementation method and system based on a storage and calculation integrated device, processing equipment and a storage medium, and the method comprises the steps: obtaining a pilot frequency structure matrix and a probability graph model of a to-be-measured channel estimation task, and determining the initialization probability of each state of each variable node in the probability graph model; the configuration of a storage and calculation all-in-one device array is carried out, and the storage and calculation all-in-one device array comprises an initial input array, an iterative calculation array and a result output array; the initialization probability serves as input of the initial input array, the iterative calculation array is used for completing iterative calculation of mutual information transmission of variable nodes and check nodes in a probability graph model of a to-be-tested channel estimation task, and the result output array is used for outputting a processing result when an iteration stop condition is met; according to the processing result output by the result output array, continuous channel estimation results in the to-be-tested channel estimation task are determined, and the method and the device can be widely applied to the technical field of signal processing.

The invention discloses a family intention recognition service method and device, electronic equipment and a storage medium, and relates to the technical field of data process.According to the family intention recognition service method and device, due to the fact that multi-modal data can be collected and real-time feature vectors can be extracted, a probabilistic graph model containing the member interaction and environmental factor coupling relation is dynamically updated; in addition, the method can also calculate a home state evaluation index and plan a service gradient sequence containing service actions and intervention intensity, thereby realizing prospective deduction of home state evolution, avoiding limitation of a static model and a preset rule, and therefore, improving the user experience. The technical problems that in the prior art, a static model is adopted, member interaction and environment coupling are not considered, a preset rule is relied on, forward-looking deduction is lacked, and dynamic requirements of complex family scenes are difficult to deal with can be solved. The technical effects of improving the adaptability of the smart home system to the dynamic demand of the complex home scene, enhancing the perspectiveness and accuracy of the service, optimizing the service intervention effect, and improving the intelligent and personalized service level of the system are achieved.

The application relates to a channel estimation method and system based on a memory-computing integrated device, a processing device and a storage medium, which comprises the following steps: acquiring a pilot structure matrix and a probabilistic graph model of a to-be-tested channel estimation task, and determining the initial probability of the state value of each variable node in the probabilistic graph model according to a pilot signal observed by a receiving end and a variance and of channel uniform quantization; constructing a node state matrix corresponding to each undirected edge in the probabilistic graph model; constructing a driving vector corresponding to each undirected edge in the probabilistic graph model; performing memory-computing integrated device configuration based on the node state matrix corresponding to each undirected edge, so as to obtain a memory-computing integrated device subarray corresponding to each undirected edge; and determining a continuous channel estimation result in the to-be-tested channel estimation task according to the memory-computing integrated device subarray, the driving vector and the confidence of message transmission between each variable node and each check node connected through the undirected edge. The application can be widely used in the technical field of signal processing.

The invention relates to the field of signal processing, and discloses a unified probability graph calculation architecture power consumption optimization method and system, a medium and equipment, and the method comprises the steps: carrying out the sparse optimization of a message passing data flow through the dynamic regulation and control of the message passing data flow in a probability iterative calculation process for a multi-task unified probability graph; and based on the optimized message passing data stream, dynamically selecting a storage and calculation integrated array node activation state through a calculation circuit of a storage and calculation integrated characteristic device so as to improve the adaptation degree of a calculation framework and hardware resources. According to the method, a message transmission data stream in a probability iterative calculation process is dynamically regulated and controlled, and a probability message with low reasoning contribution is deleted so as to reduce a calculation-storage scale; and the activation state of the storage and calculation integrated array node is dynamically selected, the adaptation degree of the calculation architecture and hardware resources is improved, and the power consumption is further reduced.

The invention relates to the technical field of robot path planning, in particular to a robot path planning method combining a black-wing optimization algorithm and a neural network, and the method comprises the steps: converting original sensor data into a standardized grid map; mapping from environment features to passing probabilities is learned through a convolutional auto-encoder, and a safety probability graph for guiding path optimization is generated; searching a global optimal path control point sequence in a solution space by combining a black wing plinual behavior model and probabilistic graph guidance; and converting the discrete path into an executable smooth trajectory through B-spline fitting and speed-curvature constraint optimization. According to the robot path planning method combining the black wing optimization algorithm and the neural network, perception-decision closed-loop fusion is realized through dynamic environment modeling and neural network guidance, and algorithm-motion depth coupling is realized through multi-target path optimization and a dynamics feasible trajectory; and dynamic scene real-time response is realized through a hierarchical re-planning mechanism and incremental updating of the probability graph.

The application discloses a circuit implementation method and device based on a unified probability graph calculation architecture, equipment and a medium, comprising: establishing a unified signal model based on the common characteristics of baseband signal processing tasks; expressing the unified signal model as a unified probability graph model; expressing the message passing from check nodes to variable nodes in the unified probability graph model as a unified signal processor architecture; pre-storing a node state matrix into a memory-computing integrated device to obtain a memory-computing operator array, driving a vector as an input analog quantity of the memory-computing operator array, and a state value probability vector as an analog quantity output by the memory-computing operator array; updating the analog quantity output by the memory-computing operator array to obtain the message passing from variable nodes to check nodes in the unified probability graph model, and inputting the same as the input of the memory-computing operator array again to enter iteration; completing iterative calculation based on a set iteration condition, and realizing a baseband signal processing task.

The invention discloses a refined disassembly method and system for a scraped car engine, and relates to the technical field of scraped car recycling. The method comprises the steps of obtaining visual data of a scrapped engine, and identifying a geometric structure, a connection relation and a surface state of the engine based on the visual data; according to the geometric structure, the connection relation and the surface state, reasoning is conducted through a probabilistic graph model, and a personalized disassembly strategy for the engine is generated; performing multi-objective optimization on the personalized disassembly strategy in a virtual environment, and planning a motion track of a robot executing the strategy; and the robot is controlled to execute the optimized motion trail, disassembling operation is completed, and real-time adjustment is conducted through force feedback in the process. Through multi-mode perception and intelligent decision making, one-engine-one-strategy personalized disassembly of the engine is achieved, value loss caused by violent breaking is avoided, high-purity secondary aluminum and remanufactured parts can be stably produced, and conversion from material degradation to value regeneration is achieved.