97 results about "Constraint matrix" patented technology

The invention relates to the technical field of oil and gas equipment operation control, in particular to an oil and gas equipment operation parameter self-adaptive fine tuning method and system, and the method comprises the steps: constructing an operation parameter optimization library driven by reinforcement learning, defining an action space containing equipment types, working condition modes and key parameter combinations, predicting accuracy, response efficiency and safety compliance are optimized in combination with a multi-target reward function, an optimized instruction set is generated, safety perception mixing precision fine adjustment is carried out in parallel, calculation precision is configured in a layered mode, errors are monitored in real time, a high-precision mode is switched if the precision exceeds the limit, and a reservoir engineering equation mechanism constraint matrix is embedded to guide parameter updating. A generative adversarial network is utilized to synthesize fault data, a Darcy seepage causal engine is integrated to correct a loss function, the system comprises an action space construction unit, a precision dynamic scheduling unit, a mechanism constraint embedding unit and a data causal cooperation unit, and the method and the system realize data-mechanism fusion, balance precision and real-time performance, and improve equipment operation safety and adaptability.

The invention belongs to the technical field of artificial intelligence, and discloses an artificial intelligence model training resource adaptive distribution system. The method comprises the following steps: acquiring and calculating graph structure data and hardware resource state data in real time, and calculating a data reuse rate and generating a candidate operator fusion scheme by constructing an operator execution time sequence constraint matrix and identifying an operator cluster of data locality characteristics; a resource competition hotspot prediction mechanism is introduced, memory bandwidth occupation fluctuation characteristics are analyzed, a resource conflict probability is calculated for a fusion scheme, and a dynamic balance optimization model of fusion income and resource conflicts is constructed. An optimal operator fusion decision sequence and a resource allocation strategy are generated through iterative solution, and accurate dynamic adjustment of computing resources in the training process is achieved. The training efficiency and the resource utilization rate are improved, the energy consumption is reduced, and the system stability is enhanced.

The invention discloses a high boundary sensitivity three-dimensional inversion method based on an electromagnetic gradient constraint matrix, and belongs to the technical field of electromagnetic signal inversion, and the method comprises the steps: constructing an initial inversion model and a target function, and introducing a model roughness item based on electromagnetic gradient constraint into the target function; calculating the electromagnetic gradient anomaly response at each measuring point under each measuring frequency, obtaining an electromagnetic gradient anomaly response vector of each measuring point, performing normalization processing, combining into a matrix, obtaining an overall normalized observation gradient anomaly matrix, and constructing an electromagnetic gradient constraint matrix in a data space; converting the electromagnetic gradient constraint matrix in the data space into a model space through a frequency-depth weighted mapping matrix; and solving the target function by adopting an iterative optimization algorithm until the model converges. According to the method, the gradient constraint matrix based on the electromagnetic field is introduced, the dynamic adjustment of the change amplitude weight of each grid unit in the roughness item of the model is realized, and the target boundary identification precision is improved.

The invention discloses a risk knowledge graph construction and intelligent early warning method based on artificial intelligence, and the method comprises the following steps: collecting and preprocessing multi-source risk related data, and generating a composite input data set; constructing an enhanced TabPFN structure and introducing a risk logic rule constraint matrix to generate a risk prediction result set; mapping the risk prediction result set to a knowledge graph entity node and relation set to generate a risk knowledge graph structure; the method comprises the following steps: introducing Sheaf Neural Networks into a risk knowledge graph structure, and generating a consistency parameter set; the knowledge graph structure is updated, and risk knowledge graph enhanced representation is generated; and performing reasoning in combination with the risk prediction result set and the consistency parameter set, and outputting an intelligent early warning result set. According to the method, table data modeling and graph consistency propagation are fused, and the risk prediction accuracy and the early warning reliability are improved.

The embodiment of the invention discloses a monocular depth estimation method and device, electronic equipment and a storage medium, and relates to the technical field of multi-modal computer vision and natural language processing.The method comprises the steps that multi-scale visual features are extracted through a visual encoder, and scene semantic description is generated in combination with image and text description; semantic categories in scene semantic description are extracted through a large language model, and a semantic prototype is formed and mapped into a high-dimensional feature vector to serve as a semantic center; calculating the visual similarity between each pixel and the visual center and the semantic matching degree between each pixel and the semantic center, performing weighted fusion to obtain a pixel group, and mapping an object space relation in the RGB image into a structured constraint matrix according to scene semantic description; and dynamically fusing visual, semantic and geometric information by adopting a cross-modal attention mechanism to generate a depth map. According to the method, the problems of insufficient semantic understanding, lack of structural constraint and superficial layer of multi-modal fusion in the prior art are solved, and the depth estimation precision is remarkably improved.

The invention provides an AI-based old community reconstruction terrain simulation display method and system, and relates to the field of terrain simulation, and the method comprises the steps: obtaining the multi-source high-precision terrain data of an old community, and carrying out the cleaning and fusion; according to the topographic data set, identifying topographic key features and classifying land cover types in combination with a semantic segmentation technology so as to obtain a digital topographic feature map with semantic tags; resident investigation and survey data, policy specifications and economic cost constraints are obtained, constraint conditions are quantified, and a structured demand constraint matrix is generated; constructing a multi-objective optimization model, and iteratively generating and verifying a terrain reconstruction scheme set by adopting deep reinforcement learning; an immersive terrain simulation environment is constructed, a scheme is adjusted in real time, user behavior data is recorded, and a final optimization scheme is output; and performing visual achievement and dynamic demonstration to obtain a multi-terminal compatible transformation scheme package. According to the method, the defect that intelligent tool application in the prior art flows in the form and cannot be deeply combined with a core transformation link is overcome.

The invention relates to the crossing field of artificial intelligence technology and bioinformatics, and discloses a pig feed efficiency prediction model and system based on multi-omics data. Modulating a neural differential equation which runs on a priori knowledge graph and is realized by a graph neural network by using the matrix so as to solve and generate a continuous evolution trajectory of an individual physiological state; and finally, aggregating the tracks, combining the constraint matrix, and outputting a feed efficiency prediction value through a second preset model. The invention further provides a corresponding prediction system which comprises a static constraint module, a dynamic core module and a prediction module. According to the method, static genetic constraint and dynamic physiological process simulation are combined, genetic differences among different individuals can be reflected, and the biological consistency and individualization precision of a prediction model are improved.

The invention relates to the technical field of satellite internet, and discloses a multi-satellite cooperative beam hopping resource allocation method and system based on interference avoidance, and the method comprises the steps: 1, constructing a multi-satellite beam hopping cooperative coverage interference analysis model; 2, modeling is carried out on a multi-satellite-hopping-beam multi-dimensional resource scheduling problem; 3, defining interference between spatial isolation measurement hopping beam patterns, constructing an interference constraint matrix, determining a shortest distance threshold value through a carrier-to-interference ratio threshold value, deducing an interference model in a same-rail scene and a different-rail scene, and bringing the carrier-to-interference ratio threshold value into optimization problem constraint; step 4, carrying out graph structure modeling based on GAT-PPO; 5, an Actor decision layer outputs a beam scheduling probability matrix and a power distribution vector in a double-branch mode, an action space is subjected to multiple constraints, and a reward function balances throughput and interference suppression; step 6, adopting a centralized training-distributed execution framework; according to the method, interference model analysis of multi-satellite cooperative coverage based on interference avoidance is constructed, and real-time analysis and scheduling of the interference condition are carried out.

The invention discloses a semi-supervised image clustering method based on adaptive graph learning, and the method comprises the steps: iteratively updating a sparse representation matrix and a paired constraint matrix until the value of a target function is unchanged or reaches a maximum iteration number; taking the updated pairwise constraint matrix as an input similar matrix, calling a spectral clustering algorithm to divide image sample data into a plurality of sample groups to finish clustering output; wherein the objective function is a weighted sum of a propagation consistency error based on a sparse representation matrix and a paired constraint matrix, an image sample data reconstruction error based on the sparse representation matrix, an L1 norm of the sparse representation matrix and a matrix correlation error based on the sparse representation matrix and the paired constraint matrix. The invention further discloses a system, a computer storage medium and a program for implementing the method. According to the method, the problem of separation of similar graph learning and constraint propagation can be solved, and the image clustering performance is improved.

The invention provides a reactor nuclide density prediction method, a reactor nuclide density prediction system and electronic equipment. The method comprises the following steps: constructing a snapshot matrix of a reactor nuclide number density sequence; using a power function to constrain nuclides in the snapshot matrix to obtain a constraint matrix; wherein the index of the power function is a score; performing DMD prediction on the constraint matrix to obtain nuclide density with constraint; and reducing the nuclide density with constraints according to the power function, and determining the nuclide density according to a reduction result. According to the method, physical driving calculation and data driving modeling are combined, a constraint function related to nuclide behaviors is introduced into nuclide evolution snapshot data, and through modal extraction and inverse transformation recovery, the capturing capability of future nuclide behaviors is improved, and the prediction precision of the number density of reactor nuclides is improved. The DMD method based on the snapshot constraint can more accurately predict the nuclide number density evolution trend at the future moment, and has the advantages of stronger generalization ability and precision.

The invention discloses an energy supply side optimization management system and an implementation method, and relates to the technical field of energy management. The method comprises the following steps: acquiring source end data of an energy supply side, analyzing the source end data through a multi-level protocol adaptation engine and a dynamic rule matching engine, and extracting various energy features and energy supply side GIS (Geographic Information System) information; constructing a multi-energy flow topology constraint matrix based on energy features, performing fine-grained geographic grid division based on energy supply side GIS information, and determining grid-level energy dynamic parameters; performing dynamic trend adjustment on the grid-level energy dynamic parameters in combination with the multi-energy-flow topology constraint matrix, and determining a grid-level energy dynamic prediction map; searching an optimal tradeoff solution set through a population aggregation and dispersion search algorithm; and carrying out constraint solution on the optimal tradeoff solution set in combination with the safe operation boundary, and determining an energy scheduling optimal solution. According to the invention, the accuracy and foresight of energy management are improved, the loss link is accurately traced, and the system is driven to continuously evolve towards the low-carbon and high-efficiency direction.

This invention discloses an environmental anomaly broadcasting method and system based on semantic analysis and knowledge graph, belonging to the field of environmental anomaly detection technology. Key technical points include: acquiring the current inspection record text and performing semantic analysis to obtain text keywords; determining the nodes corresponding to the text keywords in a preset knowledge graph based on a conditional random field (CRF) model; training the CRF model according to a preset constraint matrix; the types of nodes in the knowledge graph include enterprises, emission outlets, processes, pollutants, and sensitive targets; obtaining anomaly information corresponding to the current inspection record text based on the association relationship between the nodes corresponding to the text keywords and nodes of each type in the knowledge graph; and broadcasting the anomaly information. This invention improves the accuracy of node positioning and generates more accurate anomaly information by constructing a constraint matrix to clarify the matching rules between text and nodes and training the CRF model accordingly.

An approach of the present disclosure includes an algorithm to encode an instruction set of a domain-specific processor automatically in an optimal way, and which yields area-effective hardware. Instruction operands and opcodes can be encoded separately or at the same time. Instructions are not encoded one by one but can be encoded in groups, and instructions are grouped automatically. The decoding logic of control signals for each group of instructions, such as register read enable and write enable, is fed to a logic minimizer to obtain an encoding constraint matrix, and the encoding of each instruction in each instruction group is done using an optimal state assignment method.

The application provides a personalized glasses frame design method and system based on 3D facial features. The method obtains facial 3D geometric data through a scanning device; extracts key feature points using an improved HRNet architecture; inputs the feature points into a multi-layer perception machine model to generate an initial frame parameter set containing 32 core parameters of four parts, i.e., frame geometry, nose bridge, temple, and overall design; constructs a multi-dimensional constraint matrix composed of five sub-matrices, i.e., geometry, style, manufacturing, comfort, and medical; establishes a constraint optimization objective function, taking the initial parameter set as the decision variable and the multi-dimensional constraint matrix as the constraint condition, and iteratively solves the optimal frame parameter set with the least constraint condition violation degree through a sequential quadratic programming algorithm; and finally generates a 3D model of the personalized frame. The application realizes full-process automatic design, ensures high adaptation of the frame to the patient's face, and is especially suitable for medical rehabilitation scenarios such as facial asymmetry or postoperative repair.

The invention discloses an online education consultation system and method based on big data, and relates to the technical field of intelligent education, and the method comprises the steps: analyzing a real-time consultation session audio and video stream, generating an emotion signal and a cognitive load pause ratio, and carrying out the feature level fusion with a user behavior feature vector, constructing an educational causal graph through a Do-Calculus algorithm, converting the educational causal graph into a root cause coding report, and generating a root cause knowledge point dependence path set; encoding the root cause knowledge point dependent path set into a quantum bit state, and generating an anti-interference learning path sequence through a quantum annealing algorithm based on a quantum constraint matrix converted by a root cause knowledge point ID; and analyzing the anti-interference learning path sequence, extracting knowledge point state and sequence data, generating a structured teaching suggestion, and converting the structured teaching suggestion into an executable consultation report. According to the method, the pertinence, intelligence and intervention effect of educational consultation are remarkably improved, and the method is widely applied to online learning platforms and personalized tutoring scenes.

The invention discloses a hydroelectric generating set overhauling and disassembling method based on an improved seat head whale migration algorithm, and aims to solve the problems of long period, high cost, poor safety and the like caused by complex structure, narrow space and dependence on artificial experience in large-scale hydroelectric generating set overhauling. The method comprises the following steps: constructing a three-dimensional visual model containing equipment and tools by using SolidWorks according to a ratio of 1: 1; constructing a geometric constraint matrix and a working area conflict matrix based on an octree to form a comprehensive priority constraint model; defining a disassembly time sequence and an operator allocation decision variable, and establishing a mathematical programming model for minimizing the total disassembly time; designing a hybrid coding mode, and proposing an improved seat head whale migration algorithm containing a self-adaptive strategy and a local search mechanism; and automatically solving an optimal disassembling sequence and a personnel allocation scheme. Verification shows that the solving quality and stability of the method are remarkably superior to those of a genetic algorithm and a particle swarm algorithm, the maintenance efficiency and safety are greatly improved, the operation and maintenance cost is reduced, and technical support is provided for stable operation of the hydroelectric generating set.

This application provides a fault propagation prediction method, apparatus, and related equipment, belonging to the field of information technology. An embodiment of this application provides a fault propagation prediction method, comprising: acquiring real-time operation and maintenance data of multiple nodes and historical operation and maintenance data within a preset first time range; constructing a service causal graph based on the historical operation and maintenance data and a preset topology constraint matrix, wherein the topology constraint matrix is ​​used to determine potential causal relationships between the multiple nodes; adjusting the edge weights of the causal edges of the faulty nodes according to the real-time operation and maintenance data to obtain an updated service causal graph, wherein the faulty nodes are one or more nodes among the multiple nodes that have experienced a fault; and predicting the fault propagation probability of the fault among the multiple nodes based on the updated service causal graph. This fault propagation prediction method can improve the accuracy of fault propagation prediction.

This invention relates to the field of digital twin technology, specifically to a method and system for monitoring the operating status of an electric arc furnace based on digital twins. The method includes: collecting multi-dimensional operating parameters and obtaining historical time series and historical normal time series; obtaining abnormal indicator sequences; constructing an initial digital twin model and obtaining a trained random forest model using first and second sample sets; statistically analyzing the sets of correct and incorrect classification trees corresponding to the triggers to obtain a feature importance weight matrix; obtaining a constraint matrix using the recognition accuracy; calibrating the initial digital twin model using the constraint matrix; inputting the abnormal indicator sequences into the random forest model to determine candidate triggers; inputting the initial operating conditions into the calibrated digital twin model to obtain simulation prediction parameter values ​​for the candidate triggers; and comparing the simulation prediction parameter values ​​with the multi-dimensional operating parameters to determine the core influencing factors. This invention can accurately determine the core influencing factors of the current state of an electric arc furnace.

The invention discloses an environmental anomaly broadcasting method and system based on semantic analysis and a knowledge graph, and relates to the technical field of environmental anomaly detection.The technical scheme is characterized by comprising the steps that a current inspection record text is obtained and subjected to semantic analysis, and text keywords are obtained; determining a node corresponding to the text keyword in a preset knowledge graph according to a conditional random field model; the conditional random field model is obtained by training according to a preset constraint matrix; the types of nodes in the knowledge graph comprise enterprises, discharge ports, processes, pollutants and sensitive targets; obtaining abnormal information corresponding to the current inspection record text according to the association relationship between the node corresponding to the text keyword and each type of node in the knowledge graph; according to the method, the matching rule between the text and the node is defined by constructing the constraint matrix, and the conditional random field model is trained according to the matching rule, so that the accuracy of node positioning is improved, and more accurate abnormal information is generated.

The invention discloses a text graph method based on concept erasure, which comprises the following steps: acquiring a text, and inputting a pre-trained text graph diffusion model to generate an original image; defining an erasure set and a retention set based on the original image; an enhanced anchoring concept matrix is constructed for the erasure set through a non-erasure feature protection module, and non-erasure features are reserved; a constraint matrix is generated for the erasure set by adding an invariant constraint module, and semantic consistency of non-erasure content is ensured; constructing an optimization objective function based on the enhanced anchoring concept matrix and the constraint matrix; and solving the optimization objective function by adopting a Lagrange multiplier method to obtain a model parameter update quantity, and generating updated model parameters so as to generate an image. According to the method, global feature consistency is protected through the non-erasure feature protection module, key semantics are reserved by enhancing invariant constraints, balance between accurate erasure of target concepts and non-erasure content integrity is achieved, and meanwhile global style consistency, non-target concept semantic coherence and reservation of main content are guaranteed.

The invention relates to the technical field of channel coding and decoding. The invention discloses a parity check polarization code construction method and device, a terminal and a storage medium. The parity check polarization code with error detection capability can be generated. The construction method of the parity check polarization code comprises the following steps: acquiring a polarization code constraint matrix; constructing an information set based on the polarization code constraint matrix, constructing a constraint set based on the polarization code constraint matrix, constructing a frozen set based on the constraint set, and constructing a verification set based on the constraint set; constructing a parity check polarization code based on the information set, the frozen set and the check set; in the decoding process of the parity check polar code, parity check bits in the check set are used as information bits, and check is carried out according to a linear constraint relationship between the parity check bits and the information bits in the information set after decoding, so that effective code words are obtained.

The invention discloses a virtual power plant demand response clearing method and system based on a ciphertext grading constraint, and relates to the technical field of virtual power plant demand response clearing, and the method comprises the steps: carrying out the ciphertext packaging of a distributed resource node scheduling related parameter of a virtual power plant, and obtaining a node scheduling ciphertext parameter; scheduling control factors are obtained according to the node scheduling ciphertext parameters and the current demand gap parameters of the power grid, node priorities are determined and layered based on the scheduling control factors, and a ciphertext grading constraint matrix is constructed; according to the ciphertext grading constraint matrix and the demand response instruction, outputting an encrypted clearing result according to a power distribution objective function; and according to the ciphertext grading constraint matrix, a dynamic threshold reconstruction set is output according to a preset key reconstruction safety threshold, key reconstruction is completed based on the dynamic threshold reconstruction set, the encrypted clearing result is decrypted through the reconstructed key, and a scheduling instruction is generated. The method has the beneficial effects that the safety and the flexibility of the clearing calculation process are considered.

The invention discloses a symbolized polyphonic data anomaly detection and error correction method based on a heterogeneous graph neural network and related equipment. The method comprises the following steps: analyzing a symbolized polyphonic data stream, and extracting a multi-dimensional acoustic feature vector; a heterogeneous topological graph G = (V, Et, Ec) is constructed, Et is a horizontal directed edge, and Ec is a vertical undirected edge; extracting and fusing space-time features through a space-time diagram convolutional network; calculating a matching degree weight w based on the statistical constraint matrix, and positioning a topological distortion node cluster through threshold judgment and a connected component algorithm; and generating a candidate node set for the distorted node cluster, optimizing and correcting errors by using a global cost function, and outputting a reconstructed sequence. According to the method, the structural sensitivity and robustness of the abnormal positioning of the polyphonic data are improved, the error correction efficiency is improved, and the method can be widely applied to scenes such as music education, creation assistance and music score editing.

The invention relates to the technical field of time-domain electromagnetic geological exploration, in particular to a time-domain superconducting electromagnetic multi-effect multi-parameter extraction method and system, and the method comprises the steps: calculating time-domain superconducting electromagnetic induction-polarization-magnetization multi-effect response, and constructing a data error fitting term based on the multi-effect response and the residual error of measured data; constructing a constraint matrix of multi-parameter correlation coefficients including resistivity, polarizability and magnetic susceptibility, and constructing a multi-parameter statistical constraint term for describing statistical correlation among multiple parameters based on the constraint matrix; constructing a regularization objective function; an improved Gaussian-Newton optimization algorithm is adopted, the regularization objective function is minimized, the optimal value of the regularization objective function is iteratively solved, and the resistivity, the polarizability and the magnetic susceptibility are obtained through synchronous inversion. According to the invention, multi-parameter efficient and accurate extraction imaging in the deep mineral exploration and geological structure identification field is realized.

The application discloses a method for optimizing infrastructure network game strategy based on linear constraint, which comprises the following steps: obtaining the topology structure of the infrastructure network, determining the strategy set of the attack side and the defense side, and constructing a basic model of complex network game; determining an index representing network connectivity, calculating the income of the attack side and the defense side under various strategy profiles in the complex network game model, and thus obtaining an income matrix; adding a restriction condition to the strategy selection of the attack side and the defense side, and thus obtaining the strategy constraint set of the attack side and the defense side; converting the solution of the basic model of complex network game into the solution of a linear programming problem, obtaining a mixed Nash equilibrium solution, and thus obtaining the strategy optimization result of the attack side and the defense side. The method can greatly widen the application of complex network game research in practice by introducing the constraint matrix countermeasure into the infrastructure complex network game.

The invention provides an aggregation subject market transaction matching method for multi-dimensional target constraint, and relates to the technical field of resource allocation optimization, and the method comprises the steps: collecting transaction participation data of a plurality of aggregation subjects, carrying out the feature extraction of each transaction participation data, and obtaining an aggregation subject feature set; constructing a multi-dimensional target optimization model by taking cost, income and stability as multiple targets based on the aggregated main body feature set, and introducing output constraints, time sequence constraints and market rule constraints to form a target constraint matrix; and under the constraint of the target constraint matrix, carrying out iterative solution on the multi-dimensional target optimization model by adopting a multi-target optimization algorithm, and generating an optimal transaction matching scheme of each aggregation main body. Through the method and the device, the technical problem of poor feasibility of a transaction matching scheme caused by difficulty in processing complex multi-target conflicts due to adoption of a simplified static optimization model in the prior art is solved, and the feasibility of the aggregation subject market transaction matching scheme is improved through multi-dimensional target optimization.

The present application relates to the technical field of steel structure test, and discloses a kind of steel structure veranda health monitoring method and system, comprising: in veranda layout sensing unit and establish finite element model;Analytic node connection relationship generates static topological constraint matrix;According to environmental real-time temperature data, solve thermal deformation response using constraint matrix, and deduct thermal deformation response from total response data to obtain nonlinear residual signal;Based on sensitivity jacobian matrix, residual signal is projected to structure physical parameter space to generate response deviation vector;According to deviation distribution, determine damage location and correct model local grid density and element stiffness, the present application establishes physical mapping path by introducing geometric topological constraint, eliminates the pseudo-mapping noise generated by environmental disturbance, improves the physical authenticity of structure damage identification under complex temperature difference working condition.

The invention is suitable for the technical field of aircraft trajectory optimization, and relates to a soft landing trajectory optimization method, medium and equipment based on second-order cone preprocessing, and the method comprises the steps: constructing an SOCP problem based on a current soft landing trajectory planning problem, and obtaining an original constraint matrix of the SOCP problem; obtaining a plurality of dense sub-blocks according to the original constraint matrix, and identifying to obtain reversible sub-matrixes in the dense sub-blocks; eliminating a reversible submatrix in the original constraint matrix to obtain a pre-processed problem; and solving the pre-processed problem to obtain a pre-processed solution, and restoring the pre-processed solution to an original dimension to obtain a solution of the current soft landing trajectory planning problem. The problems of long solving time, high complexity, lack of a targeted optimization method and the like in a high-dimensional and complex constraint scene in the prior art are solved.

The application discloses a stage lighting and lifting equipment linkage control method based on timing constraint analysis, comprising the following steps: constructing a layered timing constraint matrix of lighting actions, lifting actions and safety checking actions, integrating action start-stop relationship, occupation relationship and interlocking relationship into a unified constraint expression, solving the problems of dispersed cross-equipment action relationship and unclear conflict boundary in the existing linkage control; extracting the precedence edge, parallel edge, mutual exclusion edge and recovery edge, generating an action index sequence and a candidate linkage sequence, improving the sequence and continuity of linkage arrangement; determining the to-be-executed linkage segment according to the link state and forming a linkage control sequence according to the difference priority, reducing the influence of communication fluctuation on key lifting actions; and finally, blocking or rearranging the conflict actions through index backtracking, improving the safety, stability and execution consistency of the linkage control.

A method for constructing a design concept generation network (DCGN) and a method for automatically generating a conceptual scheme are provided. A DCGN includes a Transformer encoder, a Transformer decoder, an importance constraint matrix generation module, an importance constraint embedding layer, a cross-attention (CA) layer, and an optimization module. A word importance constraint is ingeniously introduced based on an attention mechanism of a Transformer to record input word constraint information contained in a generated text sequence. This can effectively ensure the reliability and effectiveness of a generated conceptual scheme and is conducive to capturing potential semantic importance information and implementing semantic knowledge reasoning.