16 results about "D algorithm" patented technology

The invention discloses a spatial non-cooperative target form reconstruction-oriented perception planning control integration method, which comprises the following steps of: extracting local semantic features of a target component in a single-view observation image through a pre-trained semantic segmentation network, coding the local semantic features and RGB (Red, Green and Blue) information into an MLP (Markup Language Protocol) of NeRF, perceiving a target geometric form and component-level semantics, and obtaining a component-level semantic feature of the target component; the perception result is optimized along with fly-around observation, evaluation is carried out, and an uncertainty thermodynamic diagram is generated; based on the uncertainty thermodynamic diagram, a space observation value function is constructed, spacecraft dynamics and view field constraints are combined, and an initial fly-around trajectory is generated by adopting an information gain weighted three-dimensional A * algorithm and optimized in real time; and designing a trajectory tracking control law and an attitude stability control law based on a time synchronization stability theory, and controlling the spacecraft to execute the optimized fly-around trajectory. According to the invention, a perception-planning-control closed-loop execution system is realized, and a closed-loop collaborative process of perception-evaluation-planning-control-re-perception is formed.

The invention discloses a mine underground space recycling project disaster-causing key factor identification method, and relates to the technical field of underground space energy storage. The method comprises the following steps: firstly, collecting multi-source monitoring data such as displacement, microseism, stress and the like, and carrying out timestamp alignment, denoising and interpolation completion preprocessing through improved bidirectional dynamic time warping, Hampel filtering and GRU-D algorithms; converting the data to generate an initial index set containing the underground space stability state rate and the inter-field coupling strength, and screening out an optimal index subset through 20-50 generations of Bayesian optimization search and 5-fold cross validation; constructing a multi-field coupling dynamic graph based on the subset; and finally, inputting the dynamic graph into a graph neural network, outputting a node importance degree sequence through a GNN Explainer algorithm, and determining disaster-causing key factors. According to the method, multi-source data and a multi-field coupling mechanism are fused, the recognition accuracy and the model interpretability are improved, and technical support is provided for safe operation and disaster prevention and control of mine underground space recycling engineering.

The application discloses a kind of based on multi-objective optimization algorithm's advertisement putting real-time effect tracking method, advertisement putting data analysis technical field, including step one: obtaining advertisement putting real-time behavior data;Step two: constructing advertisement putting effect window matrix;Step three: through improved MiniRocket network, execute advertisement effect conduction difference feature extraction and continuous response fragment aggregation processing;Step four: constructing multi-objective tracking objective function set and multi-objective optimization constraint condition;Step five: using improved MOEA / D algorithm, execute state traction decomposition optimization and fragment duration constraint neighborhood replacement processing;Step six: by Canberra distance, execute target deviation analysis processing;Step seven: match target advertisement plan's real-time effect tracking state.The application improves the accuracy of advertisement putting real-time effect tracking by improved MiniRocket network and improved MOEA / D algorithm.

This invention discloses an intelligent optimization method for communication chip design parameters based on deep learning, comprising: S1, constructing chip topology graph data and encoding a graph structure input using Kirchhoff's laws; S2, collecting historical data to construct a standardized training sample library; S3, aggregating node information to learn parameter coupling relationships and outputting a performance prediction model; S4, introducing an improved QPSO algorithm to construct a physically constrained pseudo-gradient guiding term, using the first derivative to force particle movement in flat regions, and outputting a potential global optimal solution region; S5, using the MOEA / D algorithm to decompose a multi-objective problem and iteratively selecting non-dominated solution sets as candidate sets in parallel; S6, backfeeding a high-precision simulator for verification and incremental updates, outputting the optimal parameter combination. This invention achieves physical constraint modeling and gradient-guided optimization of design parameters, effectively improving optimization efficiency, prediction accuracy, and design feasibility.

The application discloses a steel modular building structure multi-specialty collaborative optimization method based on an improved MOEA / D algorithm and belongs to the technical field of physical calculation. In view of the problems of existing multi-specialty data fragmentation, index calculation relying on manual operation, time-consuming numerical simulation and strong subjectivity of optimal solution selection, the application firstly extracts geometric data based on drawings, fills and fuses information through professional coding, and constructs an integrated digital basis; further, building material cost is automatically calculated, structure performance indexes are extracted, and building cold and heat loads are calculated; then, isomorphic graph and heterogeneous graph data are constructed, a GCN structure agent model and a HGCN energy consumption agent model are trained; next, an optimization problem is constructed with material cost minimization and energy consumption minimization as optimization objectives and structure performance as constraints, an improved MOEA / D algorithm with a dynamic neighborhood strategy is adopted, optimization iteration is carried out in combination with the agent model, and a Pareto optimal solution set is output; and finally, an optimal design scheme is obtained through screening.

The application discloses a steel modular building structure multi-specialty collaborative optimization method based on an improved MOEA / D algorithm and belongs to the technical field of physical calculation. In view of the problems of existing multi-specialty data fragmentation, index calculation relying on manual operation, time-consuming numerical simulation and strong subjectivity of optimal solution selection, the application firstly extracts geometric data based on drawings, fills and fuses information through professional coding, and constructs an integrated digital basis; further, building material cost is automatically calculated, structure performance indexes are extracted, and building cold and heat loads are calculated; then, isomorphic graph and heterogeneous graph data are constructed, a GCN structure agent model and a HGCN energy consumption agent model are trained; next, an optimization problem is constructed with material cost minimization and energy consumption minimization as optimization objectives and structure performance as constraints, an improved MOEA / D algorithm with a dynamic neighborhood strategy is adopted, optimization iteration is carried out in combination with the agent model, and a Pareto optimal solution set is output; and finally, an optimal design scheme is obtained through screening.

The invention provides an industrial control intrusion prevention method and system based on multi-scale fusion and dynamic optimization, and relates to the technical field of industrial control system safety and deep learning crossing. According to the method, a layered closed-loop architecture of data preprocessing, feature extraction, attack detection, strategy optimization, dynamic adjustment, defense execution and feedback optimization is constructed, and shallow layer and deep layer features of a network layer, a physical layer and a cross layer are captured through a multi-scale feature fusion module; generating a Pareto optimal defense strategy set for balancing defense cost, attack loss and system availability based on an MOEA / D algorithm; millisecond-level adjustment of the strategy is realized through a Q learning dynamic optimization module; by means of a lightweight network and a closed-loop feedback mechanism, adaptive balance between high reliability and a resource-limited terminal is achieved. The method can adapt to industrial control systems in the fields of electric power, chemical engineering and the like, and effectively solves the problems of one-sided capture, single strategy optimization, dynamic response lag and unbalanced resource adaptation in the prior art.

The invention relates to the technical field of ocean observation point layout, in particular to a deep and far sea offshore observation point layout method based on multi-element fusion, comprising: based on a preset clustering algorithm, obtaining an initial clustering center corresponding to each marine environment element, the preset clustering algorithm comprising an improved GMM-p algorithm and an improved KM-D algorithm; based on preset fusion strategies, a fusion clustering center is obtained, and the preset fusion strategies comprise a gravity center fusion strategy and an average weight fusion strategy; obtaining a condensation index based on a preset condensation algorithm; and determining the fusion clustering center corresponding to the condensation combination with the minimum condensation index as the layout position of the deep and far sea offshore observation point. Through multi-element spatio-temporal data standardization, clustering algorithm improvement and multi-strategy fusion, scientization and intelligentization of deep and far sea observation point layout are realized. Introducing a condensation index, and screening an optimal layout scheme; the multi-element features of the marine environment can be comprehensively reflected, and the stability and representativeness of clustering and fusion results are improved.

This invention discloses a reentrant hybrid flow shop scheduling method based on the IMOEA / D algorithm. Specifically, it first establishes a BPM-RHFSP multi-objective optimization mathematical model with the optimization objectives of minimizing maximum completion time, total delay time, and total equipment energy cost. This transforms the actual production job scheduling problem into a combinatorial optimization mathematical model problem. Then, based on IMOEA / D, it solves the problem model by designing a variable threshold batching strategy for batch processing job decoding; multi-region global search and random variable neighborhood search; and reinforcement search for individuals within the elite solution set. This invention can effectively coordinate the processing of multiple equipment, improve the rationality of resource allocation, shorten the workshop manufacturing cycle and product delay time, and reduce energy costs.

The invention relates to the technical field of shield construction control, and discloses a shield receiving stage foam concrete adaptive parameter optimization decision-making method, which comprises the following steps: firstly, based on a multi-source signal, calling a rigid cutting characteristic reference library to execute spectrum difference cancellation operation on a time domain vibration signal, dynamically filtering background noise generated by cutting reinforced concrete, and calculating an adaptive parameter of the foam concrete; extracting a fluid damping residual error spectrum representing the rheological characteristics of the medium; thirdly, calculating a microbubble stability index according to residual spectrum characteristics and mechanical power inversion, and performing active correction in combination with a phase synchronization pulse excitation mechanism; and the system identifies the current physical rheological state according to the index, generates a control instruction by using a state-based hierarchical PI D algorithm, and drives the execution subsystem to adjust injection parameters. According to the method, the problem that fluid characteristics are covered by strong-rigidity noise in the receiving stage is solved, and real-time quantitative perception and closed-loop regulation and control of foam concrete properties are achieved.

The application discloses a computing power scheduling method of an AI server cluster based on big data mining, and comprises the following steps: S1, collecting node load and task demand data; S2, mapping a state and demand vector and calculating a resource hunger index; S3, inputting an improved MOEA / D algorithm, constructing a topological similarity dynamic neighborhood, calculating an affinity score and reducing a weight with adaptive punishment, and outputting a target node; S4, detecting resource conflicts, issuing a transfer instruction for high-priority tasks to occupy, and recording a borrowing and compensation relationship; S5, returning resources and compensation time length after the completion of high-priority tasks, integrating results, and outputting a final scheduling decision; S6, deploying and monitoring, migrating data in case of failure, counting the cost, outputting a closed-loop feedback, and adaptively correcting the weight and the penalty coefficient. The application improves the matching degree, effectively resolves resource conflicts, and realizes efficient and robust scheduling of the cluster.

This invention relates to a tax planning scheme and early warning system and method. The system includes: a regulatory data collection and verification module, which uses a blockchain-style hash chain to detect regulatory change events in real time; a time-series knowledge graph module, which stores regulatory, preferential, and scheme nodes with effective time intervals and time-series relationship edges, and executes a reverse breadth-first propagation algorithm in response to change events to calculate the degree of failure and output early warnings; an optimization modeling module, which constructs a five-objective optimization model targeting tax burden, cost, cash flow, profit fluctuation, and audit risk; a solver module, which uses an improved MOEA / D algorithm with business rules guiding the initial population and dynamically adjusts weights to solve for the Pareto optimal solution set; an output interaction module, which generates natural language explanations based on SHAP values ​​and learns user preferences through Bayesian optimization; and a visual early warning board to display the impact propagation path. This invention achieves adaptive planning updates driven by regulatory changes, ensuring the interpretability and compliance of the scheme.

The embodiment of the invention provides a multi-target multi-crown-block scheduling method based on dynamic priority conflict resolution, and belongs to the technical field of control, and the method specifically comprises the steps: 1, obtaining historical data; 2, calculating the correlation degree of each task according to historical data, dividing each task into different priorities according to the correlation degree, and designing corresponding delay penalty functions according to the different priorities; step 3, taking minimization of completion time, delay penalty and passive transportation distance as optimization objectives, and constructing a multi-objective multi-crown-block scheduling optimization model; 4, based on the dynamic priority of the time window constraint calculation task, forming a comprehensive evaluation method of conflict resolution, and establishing a simulation model according to the comprehensive evaluation method; and step 5, based on the simulation model, adopting an improved MOEA / D algorithm to solve the multi-target multi-crown-block scheduling optimization model, and generating an optimal scheduling track and a task allocation scheme of a plurality of crown blocks. Through the scheme disclosed by the invention, the scheduling efficiency and accuracy are improved.

This invention discloses an integrated collaborative modeling and bidirectional updating method and system for steel modular building structures, belonging to the field of physical computing technology. Addressing the problems of fragmented multi-disciplinary data, low optimization iteration efficiency, and reliance on manual reconstruction for output, this invention extracts geometric data from DXF drawings, fills in the code and integrates information from the architectural, structural, and energy consumption disciplines, and constructs an integrated digital foundation. Based on this foundation, it automatically calculates building material costs, structural performance indicators, and building heating and cooling loads; it constructs isomorphic and heteromorphic graph data, and trains the GCN structural proxy model and the HGCN energy consumption proxy model; with cost minimization and energy minimization as optimization objectives, it uses an improved MOEA / D algorithm incorporating a dynamic neighborhood strategy combined with the proxy model for optimization iteration; after selecting the optimal design variables, it updates the integrated digital foundation in reverse, realizing integrated collaborative modeling and bidirectional updating of steel modular building structures, and automatically mapping and outputting the IFC model.

The invention provides a gear transmission system optimization method adopting an improved MOEA / D algorithm. According to the method, constraint condition processing and a weight vector adaptive strategy of an MOEA / D algorithm are improved, a traditional MOEA / D algorithm is taken as a basic framework, simulated annealing penalty function processing is carried out on constraint conditions, an initial weight vector is generated by adopting a uniform stochastic method, and adaptive adjustment is carried out on the weight vector; the Pareto leading edge can keep uniform distribution and is prevented from falling into local optimum; the minimum acceleration, the highest reliability and the maximum power density of the gear transmission system serve as optimization objectives, design variables are screened in a partial derivative solving mode, and the system size, the material fatigue strength and the like serve as constraint conditions to establish a gear transmission system multi-objective optimization model. The problems that an NSGA-II algorithm, a genetic algorithm and the like are low in calculation efficiency and a traditional MOEA / D algorithm is prone to falling into local optimum under complex constraint conditions are solved, and better gear transmission system optimization performance is achieved.

The invention provides an unmanned aerial vehicle route planning method and system based on an improved three-dimensional A * algorithm, and the method comprises the following steps: S1, loading a three-dimensional grid map, and building an OPEN list and a CLOSED list; s2, a starting point heuristic value is calculated and added into an OPEN list; s3, the node with the minimum f value is taken out from the OPEN list to serve as the current node; s4, judging whether the current node is a target point or not, if so, backtracking the path, otherwise, entering step S5; s5, adding the current node into the CLOSED list; s6, candidate neighborhood nodes are generated based on 5 * 5 * 5 neighborhoods and turning angle limitation; s7, traversing the candidate neighborhood nodes, and updating the OPEN list; and S8, repeating the steps S3-S7 until the path is found or the OPEN list is empty. The method has the beneficial effects that the generated air route meets the maneuvering characteristic condition of the unmanned aerial vehicle, the safety is remarkably improved, the searching efficiency is effectively improved, and the practicability is higher.