14 results about "Parallel dynamics" patented technology

The invention discloses a hydraulic motor fault diagnosis method and system based on heterogeneous asynchronous data fusion, and the method comprises the steps: collecting heterogeneous asynchronous data of a sensor network, carrying out the preprocessing of the data, inputting the data into a parallel dynamic pruning residual network, extracting features, and carrying out the preliminary fusion, thereby obtaining an initial feature plane; and inputting the initial feature plane into a multi-connection neural network, fusing data, extracting features, obtaining a final feature plane, using the fused feature data as a training set, and building a feature classification model to test and evaluate the performance of the feature classification model. According to the method, the limitation of a traditional fusion model in asynchronous processing of high-frequency vibration signals and low-frequency thermodynamic data of a hydraulic system is effectively overcome, the robustness of key fault features in a strong noise environment is remarkably improved, and rapid virtual-real mapping of bench test simulation data and online monitoring data is realized; typical faults such as plunger pair abrasion and valve plate cavitation of the hydraulic motor can be accurately supported, and safety guarantee is provided for equipment life prediction and safety control.

The invention relates to the technical field of battery management systems, and particularly discloses a multi-dimensional parameter fused multi-battery-pack parallel dynamic balance control method and system, and the system comprises a multi-dimensional space-time parameter deep fusion module, an aging driving parameter weight self-adaption unit, a twin closed-loop strategy iteration system and a battery pack dynamic balance execution layer. Utilizing the state data output by multi-dimensional parameter fusion to drive aging weight adjustment; a weight strategy is verified and optimized through a twin system, and then feedback parameter fusion and weight adjustment are performed, so that the system continuously adapts to the battery state change. Multi-dimensional parameters such as voltage, current, temperature, internal resistance, SOC, SOH and the like are collected, time-space correlation characteristics are constructed in combination with timestamps and spatial position information, electrochemical, thermal and life states and time-space scene correlation of the battery are comprehensively covered through preprocessing and time-space attention mechanism neural network fusion, the one-sidedness of traditional single parameter evaluation is solved, and the evaluation accuracy is improved. Potential imbalance risks are identified in advance.

The invention discloses a rice bakanae disease bacterium-carrying seed detection method and device based on a hyperspectral image, and the method comprises the steps: preprocessing the hyperspectral data of rice seeds, and inputting the preprocessed data into a rice bakanae disease bacterium-carrying seed discrimination model; a hierarchical and adaptive feature learning and decision-making system is realized by using a multi-scale spectral feature extraction module, an adaptive spectral attention mechanism module, a deep feature cross fusion network module and a result discrimination module, the self-adaptive spectrum attention mechanism module dynamically analyzes the importance of each wave band, and the judgment result module adopts a parallel dynamic meta-learning classifier module and an uncertainty quantification module, and provides the credibility of a judgment result while outputting the judgment result of the seed health state. According to the method, efficient, accurate and lossless bacteria-carrying seed identification with quantifiable result reliability is realized.

The invention relates to the technical field of hyperspectral image processing, and discloses a parallel dynamic multi-hop graph and composite multi-scale convolution hyperspectral sparse unmixing network, which comprises a dynamic multi-hop graph interactive attention module (DMGIAM) and a composite multi-scale convolution space-spectral attention module (CMS2AM). The DMGIAM captures long-distance spatial dependency features through dynamic jump perception graph convolution and a cross-head interactive attention mechanism; the CMS2AM extracts local spectrum-space features through a multi-scale convolution residual error and a double-convolution attention module. In addition, the invention further provides a self-adaptive weighted total variation loss function (SGATV) based on a Sobel operator and a Gaussian function, noise is effectively restrained, and edge details are reserved. According to the parallel dynamic multi-hop graph and the composite multi-scale convolution hyperspectral sparse unmixing network provided by the invention, the defects of an existing method in the aspects of long-distance feature capture, multi-scale information fusion and calculation efficiency are overcome, and the precision and robustness of hyperspectral unmixing are remarkably improved.

Detection and classification of anomalies for powder bed metal additive manufacturing. Anomalies, such as recoater blade impacts, binder deposition issues, spatter generation, and some porosities, are surface-visible at each layer of the building process. A multi-scaled parallel dynamic segmentation convolutional neural network architecture provides additive manufacturing machine and imaging system agnostic pixel-wise semantic segmentation of layer-wise powder bed image data. Learned knowledge is easily transferrable between different additive manufacturing machines. The anomaly detection can be conducted in real-time and provides accurate and generalizable results.

The application discloses a kind of discrete dynamic event tree and nuclear simulation program efficient parallel dynamic coupling analysis method, this method includes: constructing discrete dynamic event tree DET simulation model, determine nuclear simulation program simulation time and run, according to nuclear simulation program simulation result analysis obtains the time information of all DET simulation object state transition control TRIP variable change, and according to DET simulation model branch rule obtains the restart time and restart number of nuclear simulation program backtracking restart of all DET simulation objects of current accident sequence, judge whether restart time and parent sequence restart time are identical, update the failure restart file as input, copy necessary restart input file in parent sequence to current folder, according to branch number backtracking executes multi-thread parallel simulation to complete the simulation of all DET failure branches.This method makes up the deficiency of traditional safety analysis method in handling the time sequence dynamic response of nuclear power plant accident process, and multi-thread parallel computing mode also improves the calculation efficiency.

The invention provides a river channel evolution model construction method and device, electronic equipment and a readable storage medium, and relates to the technical field of intelligent water conservancy. By constructing normal prior distribution of riverbed roughness parameters and combining a customized likelihood function and a Bayesian inference framework, accurate estimation of posterior distribution of the parameters is realized, the defects that a traditional method only outputs a single optimal solution, is easy to sink into local optimum and cannot quantify uncertainty are overcome, and the parameter calibration precision is remarkably improved. A self-adaptive and multi-chain parallel dynamic simulation algorithm is adopted, parallel sampling is carried out, the step length is dynamically adjusted, and the sampling efficiency and the convergence speed in a complex posterior space are greatly improved. Through deep coupling of a hydrodynamic error model and a likelihood function, the consistency of the model and an actual hydraulic process is enhanced. Parameter uncertainty is propagated to simulation results, mean values, variances and confidence intervals of water level, flow and other prediction results are constructed, and a scientific and reliable uncertainty quantification basis is provided for flood early warning and scheduling decision making.

The invention relates to the technical field of artificial intelligence monitoring, in particular to an epidemic disease early warning method and system fusing a depth probability graph model and Bayesian inference, and the method comprises the steps: inputting multi-source time sequence monitoring data and static attribute features into a depth generation model, mapping the data to an independent potential space, and decoupling and outputting pathology, environment and noise variables. The pathological variables are purified by using the two; obtaining parallel dynamic propagation structure particles based on the purification features, and inputting a continuous time evolution model to generate a plurality of epidemic situation evolution trajectories; and finally, a comprehensive risk value is calculated through a risk sensitivity evaluation function to trigger early warning. Environment drift and random noise are accurately stripped from the mixed signals, and the false alarm rate is remarkably reduced; and meanwhile, by deducing multiple propagation hypotheses in parallel and amplifying a high-risk trajectory weight by using nonlinear aggregation, it is ensured that long-tail disaster risks can be effectively captured in the face of uncertainty.

The application discloses a kind of rice seed detection method and device based on hyperspectral image of seed-borne smut, the hyperspectral data of rice seed is inputted after pretreatment rice seed-borne smut discrimination model, utilize multiscale spectral feature extraction module, adaptive spectral attention mechanism module, deep feature cross fusion network module, discrimination result module, realize hierarchical, adaptive feature learning and decision system, wherein, multiscale spectral feature extraction module realizes the extraction of different scale spectral features, adaptive spectral attention mechanism module dynamically analyzes the importance of each band, discrimination result module uses parallel dynamic meta-learning classifier module and uncertainty quantification module, while providing the credibility of the judgment result, the discrimination result of output seed health state is provided.The application realizes efficient, accurate, non-destructive, quantifiable result reliability of seed-borne smut identification.

The invention discloses an operation and maintenance command processing method and device, equipment, a storage medium and a program product, and relates to the technical field of server operation and maintenance, and the method comprises the steps: converting an operation and maintenance command into a node in an operation behavior graph, recording the incidence relation between metadata through the graph, reducing the dependence on artificial experience, and improving the efficiency of operation and maintenance. Meanwhile, the problems of configuration errors, improper command sequences and the like are avoided through structured atlas analysis; according to the risk prediction model based on the graph neural network, accurate risk probability calculation can be performed on a to-be-executed command in a dynamic and complex environment by using operation and maintenance command metadata and association relationship information in a graph, and the limitation of a traditional static rule is broken through; a safety control strategy is determined through the risk probability, real-time risk assessment and intervention suggestions can be provided before the command is executed, the problems of system interruption, data exception and the like are effectively prevented, and the method adapts to a multi-platform, multi-role and multi-task parallel dynamic operation and maintenance environment.

The invention provides a visual question-answering method and system based on dynamic balance feature space adjustment, relates to the technical field of artificial intelligence and multi-modal machine learning, and aims to solve the problems that in the prior art, a language priori problem exists, full utilization of visual information is neglected, generalization ability is reduced, and dynamic adjustment is difficult to carry out on specific samples. According to the method, related visual question and answer sample data are obtained, feature extraction is performed through an image-text joint feature extractor, extracted visual image features and question text features are fused through a multi-modal fusion module, and a supervised contrast learning mechanism and weighted fusion are adopted to obtain a question and answer feature fusion model; and respectively inputting the fusion features into parallel dynamic balance feature space branches and rare answer perception branches to respectively obtain corresponding non-normalized scores, and carrying out weighted fusion and processing to obtain a final question and answer result. The problems existing in the prior art are solved, the robustness and generalization ability of the model are improved, and the question and answer performance is improved.

Method, apparatus, and system for parallel dynamic mesh alignment are provided. The process may include determining that a temporal alignment is present between one or more frames in a received input mesh including a plurality of polygons that describe a surface of a volumetric object, then, determining an intra-frame alignment scheme to spatially align charts within a frame. The process may also include applying the intra-frame alignment scheme to one or more corresponding charts in the one or more frames for inter-frame alignment; and processing the one or more frames in parallel based on the intra-frame alignment scheme and inter-frame alignment.

The invention discloses an electric energy meter data diagnosis method and system based on dynamic resource scheduling, and belongs to the technical field of electric energy meter diagnosis. The method comprises the steps of obtaining an electric energy meter data stream; according to a preset computing power resource, performing fixed-point operation on the electric energy meter data stream to obtain an optimized electric energy meter data stream; inputting the optimized electric energy meter data stream into a pre-constructed data diagnosis model, and outputting an electric energy meter data diagnosis result; wherein the construction of the data diagnosis model comprises the following steps: adding a dynamic cutting convolution layer and a simplified residual GRU unit which are parallel to each other in front of an adaptive feature fusion layer; the dynamic clipping convolution layer is obtained by pruning the depth separable convolution layer through an entropy value of a historical optimization electric energy meter data stream, and the simplified residual GRU unit is obtained by combining an update gate and a reset gate of the GRU unit. The method is low in calculation complexity and high in efficiency, and can solve the problem of hybrid deep neural network deployment on a low-calculation-power hardware platform.

Detection and classification of anomalies for powder bed metal additive manufacturing. Anomalies, such as recoater blade impacts, binder deposition issues, spatter generation, and some porosities, are surface-visible at each layer of the building process. A multi-scaled parallel dynamic segmentation convolutional neural network architecture provides additive manufacturing machine and imaging system agnostic pixel-wise semantic segmentation of layer-wise powder bed image data. Learned knowledge is easily transferrable between different additive manufacturing machines. The anomaly detection can be conducted in real-time and provides accurate and generalizable results.