152 results about "Network embedding" patented technology

The invention relates to the technical field of geotechnical mechanics and engineering, and particularly discloses a soft soil foundation settlement monitoring method and system based on multi-field multi-source information, and the method comprises the steps: collecting multi-source data of a target soft soil area; performing constitutive parameter inversion, data standardization and discrete Fourier transform frequency domain conversion on the data to obtain a standardized frequency domain multi-field multi-source data set; based on the data set, a soft soil constitutive parameter library and a soil mechanics physical equation, constructing an FD-PINN frequency domain physical information neural network, embedding the physical equation as a prior constraint, and training the model by adopting an alternating optimization strategy; inputting the real-time frequency domain data flow into the model, and outputting the current settlement amount, the settlement rate and the multi-physical field frequency domain distribution; and in combination with a pre-established large scale model test result, through IDFT inverse transformation, a time-space domain settlement field is reconstructed, multi-stage early warning is triggered, a targeted reinforcement scheme is recommended, the soft soil foundation settlement monitoring precision and the engineering practicability are remarkably improved, and the method is suitable for construction, operation and maintenance of infrastructures such as high-speed rails and highways.

The invention discloses an entity digitization and link framework algorithm based on a heterogeneous graph attention network, and the algorithm comprises the following steps: S1, heterogeneous information network construction: carrying out the unified modeling of all multi-source heterogeneous data into a heterogeneous information network containing various types of nodes and various types of edges, s2, meta-path definition and guidance: defining "meta-paths" connecting different types of nodes to capture a complex deep semantic relationship, S3, heterogeneous graph attention network embedding: adopting an attention mechanism to enable a model to automatically learn importance of different neighbor nodes and different meta-paths, generating a final embedding vector of each entity, and establishing a heterogeneous graph attention network model; according to the method, the information fidelity is higher, modeling is directly conducted on different types of nodes and relations on a heterogeneous graph, more abundant and heterogeneous semantic information in data can be reserved compared with a multi-view method, and the end-to-end learning ability is higher; and the complexity of manually designing a fusion strategy is reduced.

The invention discloses an integrated circuit equipment data optimization monitoring system and method based on big data, and relates to the technical field of integrated circuit manufacturing. The method is used for solving the problems of insufficient multi-physical field monitoring, difficulty in abnormal traceability and lack of closed-loop control in plasma etching. A plasma sheath thickness inversion model and an etching selection ratio model are constructed by collecting radio frequency reflection phase, mass spectrum ion strength and wafer temperature data, and process parameter-plasma state dynamic coupling is established. Interference image distortion features and electron microscope size data are fused, micro-groove geometric parameters are analyzed, morphology instability risk indexes are generated, and nanoscale early warning is achieved. And designing a dual-channel fusion network, embedding a physical constraint attention mechanism, and generating an etching rate optimization instruction. On the basis of incremental learning, model parameters are updated online, a'monitoring-decision-feedback 'closed-loop system is formed, anomaly detection sensitivity and decision reliability are improved, and technical support is provided for intelligence of integrated circuit equipment.

An additive manufacturing real-time process parameter optimization method based on reinforcement learning driving comprises the steps that firstly, a multi-mode online monitoring hardware platform is constructed, and a visible light camera, a thermal imaging camera and an acoustic sensor are integrated to sense the working condition of the additive manufacturing process in real time in an omnibearing mode; secondly, extracting key features, including visible light images, thermal imaging and acoustic signal branches, of modal data on line through a lightweight CNN-Transform hybrid network, and generating a unified low-dimensional state vector through convolution feature extraction and fusion of a Transform encoder; then, a reinforcement learning model is established, a multi-target reward function is designed in combination with the extracted feature data, a process jitter penalty term, an overheating penalty term and the like are covered, and dynamic mapping of parameters and performance is achieved. And finally, an online learning and real-time decision-making system is deployed, the trained strategy network is embedded into manufacturing equipment, self-adaptive adjustment and closed-loop control of technological parameters are achieved, and the stability and product performance of the additive manufacturing process are effectively improved.

The invention discloses a multi-unmanned aerial vehicle cooperative inspection trajectory control method for optimizing medical area coverage and service efficiency. The method comprises the following steps: firstly, constructing a medical multi-unmanned aerial vehicle auxiliary inspection mobile edge computing system model, defining an unmanned aerial vehicle and mobile user set, and establishing a communication model containing A2G and A2A links, an unmanned aerial vehicle mobile model and an energy consumption model; then taking a joint function of a coverage score, a system throughput and an emergency task completion rate as an optimization target, under energy and communication connectivity constraints, proposing an LT-MADDPG algorithm, adopting a CTDE framework, processing a time sequence state by an actor network integrated with LSTM, fusing global information by a commentator network embedded with Transform through multi-head attention, and finally obtaining an emergency task. And modeling a cooperative relationship between the unmanned aerial vehicles and a medical task priority. Experiments show that the method is superior to a traditional algorithm in the aspects of coverage, service fairness and system throughput, and the medical inspection efficiency and reliability are effectively improved.

The invention discloses an artemisinin extraction intelligent tracking and identification method, and relates to the technical field of image data processing. According to the method, an integrated closed-loop system is constructed through cooperation of four core technical means: based on classified filtering of foam stability difference and bimodal adaptive segmentation, dynamic foam is accurately removed, and preliminary interface extraction is optimized; the attention enhancement U-Net network is embedded into a channel attention module and an edge enhancement branch to realize accurate positioning and fluctuation tracking of a layered interface; the attention mechanism dynamically adjusts the bimodal weight, and completes multi-index cooperative tracking in combination with a partitioning strategy and an improved network; a crystal growth model is introduced to correct particle weight, a tracking effect is optimized by adopting layered resampling, and a closed-loop feedback mechanism of a crystal state and process parameters is established. All technical means are deeply coordinated, intelligent and accurate management and control of the whole extraction process are achieved, and reliable technical support is provided for artemisinin extraction production.

The invention discloses an angle steel connecting piece shear strength prediction method and system based on a physical information neural network, electronic equipment and a storage medium thereof. The method comprises the following steps: establishing a database containing a plurality of groups of test data based on a numerical simulation result of an angle steel connecting piece finite element model verified by a push-out test; embedding an angle steel connecting piece shear bearing capacity physical constraint condition based on an elastic foundation beam theory into a loss function of the data-driven neural network model DNN; training the neural network embedded with the physical constraint condition by using a database, and adjusting and selecting a physical item weight factor and a model learning rate so as to construct a shear bearing capacity prediction model of a physical information neural network (PINN); predicting the shear strength of the angle steel connecting piece by using the trained physical information neural network (PINN) shear capacity prediction model; according to the prediction method and system, the electronic equipment and the storage medium thereof provided by the invention, the accuracy and reliability of the shear resistance prediction of the angle steel connecting piece can be improved.

The invention discloses a blasting funnel volume solving method based on an improved physical information neural network. The method comprises the following steps: step 1, establishing a mechanical model of blast hole wall blasting load; 2, establishing a blasting physical model in which a columnar cartridge bag is equivalent to a spherical cartridge bag by using a Starfied superposition method; step 3, constructing a blasting funnel volume prediction model based on the wavelet multi-scale synchronous compression transformation enhanced physical information neural network, constructing a solution space of a physical field by using the wavelet multi-scale synchronous compression transformation, and training the physical information neural network in which a blasting funnel physical control equation, an initial condition and a boundary condition are loss functions; and 4, intelligently predicting the volume of the blasting funnel by adopting the trained enhanced physical information neural network. The trained physical information neural network enhances the robustness and generalization ability of the blasting funnel volume prediction model, and provides high-precision theoretical support for blasting design optimization in engineering blasting.

The invention belongs to the technical field of multi-source fusion positioning, and particularly relates to a multi-source fusion positioning method based on LSTM-KF, a program, equipment and a storage medium. According to the method, a long short-term memory (LSTM) neural network is embedded into a Kalman filtering framework, nonlinear error compensation of system state prediction and observation updating is achieved, and then an improved Kalman filter with the dynamic noise adaptive capacity is constructed. By loosely coupling GPS satellite positioning, IMU inertial measurement and VO visual odometer multi-source heterogeneous sensing data, the millimeter-level precision of robot pose estimation in a complex dynamic environment and the anti-interference capability of the system are remarkably improved.

The invention discloses a medicine sales management method and system based on information monitoring, and relates to the technical field of medicine sales management, and the method comprises the steps: collecting medicine sales time sequence data, carrying out the cleaning and structural processing, and generating standardized sales data; based on the standardized sales data, constructing a sales behavior time sequence causal network, embedding virtual sales intervention nodes, and generating a time sequence causal graph; the intervention strategy is coded to the virtual intervention node in the time sequence causal atlas, the graph neural network is utilized to simulate time sequence propagation of the intervention strategy, and a sales prediction result is obtained; and combining the sales prediction result with the real-time inventory, constructing a multi-level distributed sales dependency network, and forming the sales dependency network. According to the method, the preset intervention strategy is embedded into the causal structure in the node form, so that dynamic modeling of the multivariable causal relationship in the sales behavior is realized, and the prediction result not only reflects the historical trend, but also can respond to the possible influence of different intervention strategies.

The invention belongs to the technical field of medical data processing, and discloses a preoperative multi-complication risk prediction method and system based on structured clinical data, and the method comprises the steps: inputting the causal association between risk factors and complication nodes into the edge of a knowledge graph, calculating the statistical correlation between all complications, and supplementing the statistical correlation into the knowledge graph, and performing network embedding training on the knowledge graph to form a first-stage model, performing preliminary risk assessment on complications, modeling the knowledge graph in a graph neural network mode, and performing joint training with the first-stage model to form a second-stage model to output a final complication probability. According to the method, the interpretability and cross-domain consistency of the model can be improved through deep fusion of the medical knowledge graph and the multi-relational graph convolutional network, stability and calibration performance are still kept in a specialist with scarce sample size, and the problem that a traditional black box model cannot be interpreted is avoided; and the practical application value can be evaluated conveniently.

The invention relates to the technical field of artificial intelligence, in particular to a reinforcement learning green large model training method based on low-carbon contribution feedback, and the method comprises the steps: constructing a knowledge vector library coded by a neural network embedded model; constructing a low-carbon contribution evaluation function by using a first neural network large language model, retrieval enhancement and expert rules, and generating a reward signal; and adopting a reinforcement learning algorithm to optimize a second neural network large language model serving as a learning agent according to the reward signal. According to the method, the dynamically quantified evaluation function is constructed as a reward signal of reinforcement learning, so that the problems that green guidance of a general model is unknown and contribution is difficult to measure are solved, and accurate molding of the specific green behavior tendency of the large model of the neural network is realized.

The method and system for data pruning use the novel heuristic of weighting the selection of images by an internal diversity metric, such as the radius of the cluster, allowing more images to be sampled from clusters that are more internally diverse. This heuristic is added to improve the overall diversity of the selected images and to prevent the over-representation of similar images. By sampling more images from clusters that are more internally diverse, the approach is able to better represent the overall distribution of the data, improving the quality of the resulting pruned dataset.

The invention relates to a road defect detection method and system based on improved YOLOv8n, and belongs to the technical field of computer vision. The problems that an existing YOLOv8n model is high in small-scale crack omission ratio in a complex road scene, the precision is insufficient under complex background interference, and irregular defects are not accurately positioned are solved. According to the method, through customized data enhancement, a C2S lightweight feature extraction module is introduced into a backbone network, a BiRatt bidirectional routing attention module is embedded into a neck network, and a Shape-IoU loss function is adopted to construct a YOLOv8-CBS model. According to the method, the detection capability of small cracks, the robustness under a complex background and the positioning precision of irregular defects are remarkably improved, and the automatic high-precision detection requirement of road maintenance is effectively met.

This invention discloses a method and system for inverting mechanical parameters and intelligently analyzing the stability of tunnel surrounding rock, relating to the field of intelligent construction technology for tunnels and underground engineering. The method includes: collecting tunnel monitoring and measurement data and constructing a displacement field observation matrix; constructing a physical information neural network embedded with the geotechnical mechanics control equations to invert the mechanical parameters and stress field of the surrounding rock; automatically calling the finite element kernel through a programming interface and calculating the safety factor of the surrounding rock using the strength reduction method; using evidence theory to fuse multi-source analysis results and output the stability level; and driving online model updates and support optimization through prediction-monitoring comparison verification. This invention achieves the integration of parameter inversion, automated numerical simulation, and closed-loop verification, improving the accuracy, efficiency, and intelligence level of surrounding rock stability analysis.

The application discloses a neural network embedding method and device for power distribution network state estimation, electronic equipment and medium, wherein the method comprises: acquiring an input sequence; embedding time information and node type information into a vector through space-time prior information embedding to obtain a space-time embedding vector; sampling a node of interest according to a power flow direction of optimal power flow and fusing node information to obtain a node embedding vector; using a graph isomorphism neural network to capture the local of a graph and embedding it into a feature vector to obtain a structure embedding vector; fusing the input sequence and the three vectors and inputting them into a graph space-time prediction network to output a prediction result. Through the introduction of space-time prior information, graph node embedding based on the optimal power flow direction and graph structure embedding of the graph isomorphism neural network, the application realizes the modeling of the characteristics of the power distribution network, makes up for the deficiency of the prior art in the specific modeling of the power distribution network and improves the accuracy of the power distribution network state estimation.

The invention provides a wind tunnel multi-target aerodynamic optimization method based on machine learning, and belongs to the technical field of wind tunnels, and the method comprises the steps: building a wind tunnel geometric parameterized model through a free deformation method, generating an initial sample through Latin hypercube sampling, and executing computational fluid dynamics simulation to obtain aerodynamic performance parameters; a physically guided deep residual network is constructed to learn a mapping relation between control point coordinates and performance parameters, and the network is embedded into a reference vector-based multi-objective evolutionary optimization algorithm as a fast fitness evaluator. A sequential sampling mechanism is triggered through a crowding degree index, a high-precision simulation sample is added in a Pareto frontier key area to continuously update an agent model, and finally an optimal control point coordinate combination which is uniformly distributed and corresponding aerodynamic performance parameters are output. The technical problem that in the wind tunnel multi-target pneumatic optimization process, the optimization efficiency is low due to the fact that the simulation calculation cost of computational fluid dynamics is high is solved.

A method includes: obtaining network entity data associated with each network entity, from each of one or more network entities; generating, using an encoder model, network embeddings for the one or more network entities, based on the network entity data; converting, using a transformation model, the network embeddings into a predefined number of parameters; inputting the predefined number of parameters to an inference model; obtaining, from the inference model, an output regarding the predefined number of parameters; and determining, based on the output of the inference model, one or more parameters associated with control of a network.

The invention discloses a transformer winding temperature field rapid calculation method based on a mechanism embedded network, and the method specifically comprises the following steps: S1, building a corresponding temperature rise full-order model according to the structure size of a transformer winding, simulating the winding temperature fields under different working conditions, and constructing a snapshot matrix; a POD order reduction method is further combined to obtain a better modal capable of representing the physical system and a corresponding modal coefficient; and S2, according to a flow-heat coupling equation of the oil-immersed transformer winding, selecting important working condition parameters influencing steady-state temperature rise of the winding, and determining a sampling range and a step length based on an actual operation working condition. Taking the determined working condition parameters as input, taking a modal coefficient solved by POD as output, and training an RBF-MLP network embedded in a modal contribution degree mechanism by adopting a training strategy combining sub-network independent training and joint training; s3, for a transformer winding temperature inversion problem under a new working condition, inputting each working condition parameter under the working condition into the trained neural network, so that a corresponding modal coefficient can be quickly mapped; and S4, carrying out linear combination on the predicted modal coefficient and the selected modal, so as to quickly reconstruct the temperature field. According to the method, the nonlinear mapping relation between the working condition parameters and the modal coefficients of the transformer is successfully fitted, and then rapid calculation of the three-dimensional steady-state temperature rise of the transformer winding is achieved.

The invention belongs to the technical field of filtering optimization, and relates to a method for constructing an interactive multi-model Kalman filtering network with unknown prior parameters, which comprises the following steps: S1, constructing a double-branch neural network which comprises a transition probability learning module and an observation covariance learning module; s2, generating training data; s3, sequentially training a transition probability learning module and an observation covariance learning module according to the training data; and S4, embedding the trained dual-branch neural network into an interactive multi-model Kalman filter, and updating the dual-branch neural network to an optimal state estimation sequence through iteration to generate an interactive multi-model Kalman filter network. Priori parameters are autonomously learned through the double-branch neural network composed of the transition probability learning module and the observation covariance learning module, so that manually preset prior parameters are replaced, and the problems of low positioning precision and poor real-time performance of robot autonomous navigation caused by the existing IMM-KF are solved.

The invention discloses a personalized course recommendation method fusing multi-view comparative learning and hierarchical feature weighting. The method comprises the following steps: data preprocessing; carrying out graph convolutional network embedding learning; performing hierarchical feature weighting, performing weighted fusion on embedding of each layer by adopting a layer attention mechanism, fusing information of different layers in a learnable manner, and finally obtaining embedded representation of the learner and the course; the method comprises the following steps: constructing multiple views, randomly injecting disturbance in the embedding of learners and courses, then designing and fusing three denoising factors, generating multiple views of the learners and the courses, and finally optimizing the views by adopting a comparative learning mechanism, so that the same learner or course is kept consistent under different views, and the influence of data noise on a model is relieved. And the robustness and generalization ability of the recommendation system are improved. According to the method, through a combined learning mode of fusing multi-view comparison and hierarchical feature weighting, the model performance can be effectively improved, the interference of data noise on recommendation is relieved, and high-quality personalized course resource recommendation is realized.

The application relates to the field of digital signal processing and artificial intelligence, and particularly discloses a floating-point exponential in-memory parallel comparison method and system for discharge timing decision, which comprises a dynamic logic controller used for generating dynamic matching timing control signals CLK0-CLK6; an exponential maximum value matching network embedded in an SRAM storage array, which comprises a plurality of matching units, and each matching unit corresponds to a row of floating-point exponential operation results; and the matching unit comprises a plurality of bit-by-bit discharge channels; through the bit-by-bit discharge mechanism from high bit to low bit, the parallel comparison of all 64-way results can be completed by only one set of dynamic nodes penetrating the array, the need for laying a plurality of groups of signal lines in the column direction is completely avoided, and the wiring congestion problem is fundamentally solved. Therefore, the operation unit can be embedded in a standard SRAM array at a very high density, and a leading storage density of 1456 Kb / mm2 is obtained.

The invention discloses an improved YOLO-based electrical equipment defect image detection method and related equipment, and the method comprises the steps: inputting an electrical equipment image into a lightweight target detection model, and outputting defect category and position information; the model is improved based on a YOLO11 architecture, and images are processed through a feature extraction network, a fusion network and a detection head in sequence. A residual error enhancement re-parameterization convolution module is embedded into the feature extraction network, multi-scale features are extracted and fused through multiple branches during training, and re-parameterization is performed into a single branch during reasoning; a self-adaptive down-sampling module is adopted to replace a stride convolution down-sampling layer, and a dual-path structure reduces the resolution and retains information at the same time; the detection head is a lightweight scale decoupling detection head, decouples target classification and bounding box regression tasks, and adopts a detail enhancement structure. The invention aims to reduce the model volume and the calculation overhead on the premise of ensuring the detection precision, realize the real-time detection of the airship airborne equipment, construct a fine defect feature retention mechanism, reduce the small target omission ratio and improve the adaptability of the model to a complex dynamic environment.

An adapter increases the resiliency and security of protocols used for communication between embedded systems on a network. The adapter actively protects against reverse engineering and exploitation. The adapter allows embedded systems to actively thwarting cyber-attacks in real-time while allowing the embedded system to continue operating in mission ready status.

The invention relates to a water environment pollution source reasoning and tracing method, system and equipment and a medium. The method comprises the following steps: constructing a pollution source characteristic database containing spectrum fingerprints and spatio-temporal information; performing time sequence enhancement on the mixed spectral signal of the monitoring point to extract stable features; a hydrological model is coupled to dynamically simulate pollutant transport paths and probabilities, and a potential contribution source set with weights is generated; analyzing the mixed signal by adopting a deep unmixing network embedded with space-time constraint, and separating and quantifying known source contribution and unknown pollution components; and finally, generating a visual pollution contribution degree spatial distribution diagram and a quantitative traceability report through geographic information mapping. The method solves the problems that complex mixed signals are difficult to accurately analyze, multi-source contribution is difficult to dynamically quantify and unknown pollution components are difficult to effectively recognize in the prior art, and intelligent and accurate traceability of municipal water environment non-point source pollution is achieved.

The invention discloses a biomolecular interaction prediction method based on multi-modal attention fusion, and belongs to the technical field of artificial intelligence drug discovery. The method comprises the following steps: acquiring multi-modal characteristics of drugs, targets, diseases and genes: sequence structure characteristics, 3D structure characteristics, similarity network characteristics and biological relation network embedding characteristics; constructing a feature fusion prediction model, and performing training; inputting the multi-modal features of the two biological entities into the trained feature fusion prediction model, and outputting the probability of interaction of the two biological entities; the feature fusion prediction model comprises a Transform encoder and an MLP (Markup Language Protocol) network; the multi-modal features are input into a feature fusion prediction model for stacking and then are input into a Transform encoder, the features are processed by using a multi-head self-attention mechanism, and an output result is flattened and subjected to dimension reduction processing to obtain embedded vector representation; and finally, performing element corresponding multiplication on the embedded vectors of the two biological entities, inputting the embedded vectors into an MLP network, and outputting an interaction probability between the two biological entities.

An intelligent concrete, self-sensing method for intelligent concrete, device and storage medium, which comprises: concrete and long-distance, large-capacity and multi-parameter optical fiber sensing cables embedded in the concrete; the intelligent concrete is used for all-round self-sensing of the external state and internal health state. The invention embeds the grating sensing network formed by the optical fiber sensing optical cable into the concrete like a neural network, enabling the intelligent concrete to have large-area and all-domain self-sensing capabilities, providing new technologies and means for intelligence in fields such as highways, airports and bridges. At the same time, the optical fiber sensing optical cable has no probes and is not prone to damage during use. Further, by setting the multi-parameter optical fiber sensing optical cable, the dimension of the measurement parameters is increased, thereby improving the accuracy of monitoring the external state and internal health state.

A method for identifying a parent company in a corporate hierarchy is provided. A neural network embeds a number of company names into respective numeric vectors and then indexes the vectors into an approximate nearest neighbors (ANN) index according to euclidean distances between them. In response to a query regarding a parent company of a specified query company, a top N number of nearest neighbor companies to the query company is extracted from the ANN index. A machine learning voting model determines which, if any, of the extracted nearest neighbor companies has a parent company that best corresponds to the query company. If an extracted nearest neighbor company has a parent company that best corresponds to the query company, the parent company is displayed to a user through a user interface.

The invention discloses a building group energy management method based on concept bottleneck and interpretable reinforcement learning. Firstly, an environment state vector of a building group is obtained, an online concept generation module adopts a random hinge forest to convert the environment state vector into a plurality of atomic concepts, concept activation values of the atomic concepts are calculated, and the concept activation values of all the atomic concepts form a concept activation vector; then, according to the concept activation vector, the interpretable action decision module adopts a soft decision tree regression model to generate equipment control actions, and regulation and control instructions representing energy storage battery charging / discharging power, cold storage tank cold storage power and hot water storage tank heat storage power; and finally, connecting the online concept generation module and the interpretable action decision module in series to form a decision model, embedding the decision model as a strategy network into the soft actor-commentator reinforcement learning framework, and training the decision model. According to the method, the transparency and interpretability of the model decision process are remarkably improved, and the control level of energy management of the building group is improved.

The present application discloses a method for reconstructing the non-uniform stress field of complex components based on boundary segmentation and frequency-domain bridging, which relates to the technical field of stress field reconstruction. The method includes: first, obtaining the geometric model of the target complex component, identifying the degree of non-uniformity of the stress distribution corresponding to the geometric boundary features and dividing it into multiple sub-regions; then, aiming at the spatial frequency characteristics of the stress distribution in each sub-region, constructing a physics-informed neural network embedded with the elastic mechanics mechanism equation, and training to obtain the stress field sub-network model of each sub-region; then, transforming the spatial-domain stress field output by the sub-network to the frequency domain, establishing a frequency-domain dynamic link between sub-regions through a frequency-domain bridging module, and reconstructing the overall non-uniform stress field of the component through inverse transformation; finally, through a variable fidelity cascaded neural operator network, gradually fusing multi-source stress field data to complete the improvement of fidelity. The method of the present application takes into account both the solution efficiency and the multi-scale stress fitting accuracy, and eliminates the boundary discontinuity problem of segmentation and splicing.