17results about How to "Reduce computing resources" patented technology

The invention belongs to the technical field of knowledge graph construction, and particularly relates to a multi-source heterogeneous data fusion and knowledge graph automatic construction system, which comprises a semantic extraction module for extracting minimum semantic fragments from multi-source data and performing cross-source alignment to generate a symbolized semantic framework; the data purification module converts multi-source data into symbolic predicates according to the framework, filters low-evidence data and outputs a purification set; the ternary generation module inputs the initial triad candidate set into a minimum rule grammar, and extracts entities, relationships and attributes to generate an initial triad candidate set; the disambiguation calibration module generates entity two-hop topological fingerprints based on the candidate set, completes disambiguation alignment and corrects conflicts, and outputs unambiguous structured knowledge; the mapping fusion module fuses the mapping information with the minimum connected ontology, and constructs and verifies an initial mapping knowledge domain; and the incremental updating module processes newly added data, performs incremental merging and maintains consistency, and forms a complete knowledge graph. According to the method, through full-link automatic construction, efficient fusion and high-quality atlas are realized.

This invention discloses a method, apparatus, device, and medium for malware detection based on subspace ensemble learning. The method includes: acquiring the installation package file of the software to be detected; parsing the installation package file to obtain the application programming interface (API) call relationships of the software to be detected; determining the API call relationship graph based on the API call relationships; filtering each API in the call relationship graph to obtain a target API set; and determining the software detection result of the software to be detected based on the target API set and a feature subspace ensemble learning model. This invention obtains the target API set by filtering the API call relationships and employs a feature subspace ensemble learning model composed of multiple machine learning algorithms, which can capture malicious behavior in software from different perspectives, thereby improving the accuracy of malware detection.

This invention provides a method for simulating buckling of subsea pipelines based on the pipeline element method, addressing the problems of low computational efficiency, complex modeling, and low accuracy in existing methods. Based on Euler-Bernoulli beam theory, this invention models the pipeline as a series of efficient pipeline elements. Internal pressure is converted into equivalent temperature rise, and the effective thermal expansion force generated by the effective temperature rise is applied to both ends of the pipeline element to construct the total potential energy equation. The second derivative of the total potential energy equation is performed to obtain the tangent stiffness matrix of the pipeline element. An updated Lagrange method is used for coordinate transformation. The secant relationship is used to calculate the pipeline element and soil resistance. A Newton-Raphson incremental-iterative numerical program is established to perform nonlinear analysis of the pipeline, progressively solving for the deformation and stress of the pipeline under complex loads. This invention significantly reduces the number of elements, improves computational efficiency, and accurately simulates nonlinear pipe-soil interactions. It is applicable to various pipelines, including short pipes, long pipes, and corroded pipes, as well as complex working conditions such as high temperature and high pressure.

The invention discloses an attention learning mechanism-based inverter model control prediction method, system and device, and a storage medium, and the method comprises the steps: pre-constructing a three-phase inverter simulation model, carrying out the real-time sampling of a state signal through employing an MPC controller, and obtaining a signal characteristic set; performing feature selection on the signal feature set, and extracting feature data; setting a screening threshold to screen the feature data, and taking the screened feature data as optimal features; inputting the optimal feature into a pre-trained machine learning model module to obtain a prediction result of the control signal of the three-phase inverter; the control performance can be improved while more calculation amount and communication traffic are reduced.

The invention relates to a non-ideal array DOA estimation method based on multi-scale weight distribution and transfer learning and electronic equipment, and the method comprises the steps: constructing a fine tuning data set under a non-ideal array condition, the fine tuning data set comprising array signal data simulating various array physical defects; constructing an improved denoising and classification network, and training the improved denoising and classification network by adopting a transfer learning strategy; inputting a test signal in a non-ideal array environment into the trained improved denoising and classification network, and performing DOA estimation; the electronic equipment is realized based on the method. According to the method, the calculation overhead is reduced, the model is more robust to the interference of array errors and noise, the feature expression ability of signals is improved, the robustness and precision of the model in a complex environment are improved, transfer learning enables the model to obtain better performance in a low signal-to-noise ratio environment through a parameter sharing mode, and the method is suitable for large-scale popularization and application. And new non-ideal array data can be quickly adapted, and the retraining time and computing resources are reduced.

The invention provides a charging behavior prediction model construction and prediction method and system based on a large model, and the method comprises the steps: carrying out the time sequence arrangement of a plurality of groups of historical charging data, selecting the historical charging data of a corresponding group number according to the window width of an incremental window method and a time sequence, and constructing a plurality of samples, and obtaining a sample set; the window width is gradually increased as time goes on; based on the sample set, training a large language model by adopting a low-rank adaptive supervised fine tuning method to obtain a charging behavior prediction model, and performing prediction by adopting the charging behavior prediction model to obtain charging data of next charging of the to-be-predicted user; according to the method and system, a sample set is constructed through multiple groups of historical charging behavior data and an incremental window method, multi-source information is fused, and the application value and prediction accuracy of a prediction model are improved; and meanwhile, historical charging behavior data of the user is deeply understood by utilizing the generation capability of the large language model to carry out comprehensive prediction, richer and more accurate prediction results are provided, and the interpretability of the prediction results is improved.

The embodiment of the invention relates to a data processing method, electronic equipment, a storage medium and photoetching equipment. The data processing method comprises the steps that for each block of a target layout, defect values of defect points in each block are sequenced according to defect types, so that sequenced defect data of each block are obtained, and the defect values indicate the severity of defects; grouping the sorted defect data of each block based on a pattern corresponding to the defect points in each block to obtain a plurality of grouped defect data of each block; filtering the plurality of grouped defect data of each block to obtain filtered defect data of each block; and obtaining full layout defect data of the target layout based on the filtered defect data. According to the embodiment of the invention, on the premise of ensuring the analysis precision, the memory and the calculation time required by full-chip defect grouping can be greatly reduced.

The application relates to the technical field of industrial anomaly detection, and discloses a rotation-invariant multi-prototype industrial anomaly detection method, which comprises the following steps: S1: obtaining an industrial image to be processed and a corresponding defect region frame-level label, and constructing a foreground mask and a background mask based on the frame-level label; S2: inputting the industrial image into a pre-trained feature extraction network, obtaining an initial feature map, and performing channel mapping and normalization processing on the initial feature map to obtain a normalized feature map; and S3: rotating the normalized feature map at least two different angles to obtain a feature map in each rotation direction. The rotation-invariant multi-prototype industrial anomaly detection method is based on a feature extraction network pre-trained on a large-scale data set, can make a heat map generation module converge only through a small number of training rounds, does not need to train a complex model from the beginning, and greatly reduces data demand, computing resource and training time cost.

The present application belongs to the technical field of electrochemical catalysis, and provides a method for predicting the catalytic performance of tubular iron-nitrogen-carbon for CO2RR with different curvatures, which is mainly used for predicting the influence of different curvatures on the catalytic performance of tubular iron-nitrogen-carbon for CO2RR, and the main scheme comprises the following steps: a half-pipe iron-nitrogen-carbon structure model is established to replace the whole-pipe structure model to preliminarily screen the curvatures, and the feasibility and advantages of this method are explored; based on the screened iron-nitrogen-carbons with different curvatures, the influence of the curvatures on the catalytic performance of the iron-nitrogen-carbon for CO2RR is analyzed; and the influence of the curvatures on the electronic properties of the iron-nitrogen-carbon system is analyzed based on Bader charge. Under the premise of ensuring the accuracy of the method, the present application saves the calculation resources, predicts the influence of different curvatures on the catalytic performance of the tubular iron-nitrogen-carbon for CO2RR, explores the influence law of the curvatures on the catalytic activity and selectivity of the iron-nitrogen-carbon catalyst, and can be used for guiding the subsequent experimental research work.

The invention discloses a breath sound classification detection method based on an improved convolutional neural network, and the method comprises the steps: firstly carrying out the preprocessing and feature extraction of a breath sound signal collected in real time, and constructing a breath sound cepstrum feature matrix; and then the breath sound features are input into an improved convolutional neural network comprising a convolution module, a channel attention mechanism module and a lightweight feature extraction module for training and classification, and accurate recognition of breath sound categories is realized. The method realizes real-time, efficient and high-precision classification detection of breath sound, has the advantages of low calculation complexity, strong real-time performance and high identification accuracy, and is suitable for the field of biomedical signal processing.

This invention discloses a hierarchical multi-label attribution method and system that integrates atomic rule-driven credible features and knowledge distillation, belonging to the field of natural language processing technology. First, this invention constructs an atomic rule base for weakly supervised text annotation. Then, it uses a large language model as a teacher model to correct and supplement the weak annotation results, extracting the probability distribution of soft labels and intermediate layer feature representations on each level of labels. Next, it evaluates the credibility of the teacher model's output, selecting a subset of credible soft labels and credible feature dimensions. Then, it constructs a student model with a hierarchical output structure, designs a joint loss function, and distills the student model for training. Finally, it deploys only the student model for inference, outputting hierarchical multi-label attribution results and key evidence fragments. This invention, through the combination of atomic rules and credible knowledge distillation, significantly reduces inference costs while improving the accuracy, stability, and interpretability of hierarchical multi-label attribution.

The application provides an automatic data governance method and system based on a multi-modal large model, comprising: collecting multi-source heterogeneous industrial data and performing standardization processing to form standardized multivariate time series data; constructing a process knowledge base, performing semantic embedding coding on process knowledge text, and storing; constructing and fine-tuning a KTSF multi-modal large model, fusing process knowledge semantics and multivariate time series data through a cross-modal attention mechanism to generate joint semantic representation; based on the prediction of the KTSF multi-modal large model, outputting the residual error between the actual data, dynamically identifying abnormal data; and performing attribution analysis; based on the attribution result, calling the KTSF multi-modal large model to generate a repair value, and intelligently correcting the abnormal data; designing a quality evaluation and feedback learning module for calculating data quality scores and driving model incremental updating; designing a rule self-learning module for automatically refining governance rules through cluster analysis and updating the knowledge base.

This invention discloses a hierarchical multi-label attribution method and system that integrates atomic rule-driven credible features and knowledge distillation, belonging to the field of natural language processing technology. First, this invention constructs an atomic rule base for weakly supervised text annotation. Then, it uses a large language model as a teacher model to correct and supplement the weak annotation results, extracting the probability distribution of soft labels and intermediate layer feature representations on each level of labels. Next, it evaluates the credibility of the teacher model's output, selecting a subset of credible soft labels and credible feature dimensions. Then, it constructs a student model with a hierarchical output structure, designs a joint loss function, and distills the student model for training. Finally, it deploys only the student model for inference, outputting hierarchical multi-label attribution results and key evidence fragments. This invention, through the combination of atomic rules and credible knowledge distillation, significantly reduces inference costs while improving the accuracy, stability, and interpretability of hierarchical multi-label attribution.

The invention discloses a method for a communication module to lightly and safely receive a cloud AT instruction and related equipment. The method for the communication module to lightly and safely receive the cloud AT instruction comprises the steps of dynamically determining whether to use a security protocol to communicate with a communication operation maintenance platform or not based on the type of the AT instruction issued by the communication operation maintenance platform; or deciding whether to use a security protocol to communicate with the communication operation maintenance platform or not based on the network environment. According to the invention, when the AT instruction received by the communication module is a non-control instruction or the communication module accesses the communication operation maintenance platform through the intranet IP address, the communication module does not use a security protocol to communicate with the communication operation maintenance platform; according to the invention, handshake during establishment of secure connection by using a security protocol and processing duration and computing resources required by encryption and decryption operation in a data transmission process are reduced, and communication time delay and power consumption of the Internet of Things terminal are remarkably reduced while communication security is ensured.

The application provides a generator control method and device, computer equipment and a readable storage medium. The method comprises the following steps: obtaining an out-of-limit condition matrix of a specified line set; determining an out-of-limit line in the specified line set when each power transmission line fails based on the out-of-limit condition matrix; determining a target generator related to current supply of the out-of-limit line in a generator set corresponding to the specified line set; determining a power regulation upper limit set of the generator set based on a power regulation upper limit value of the target generator; generating an output constraint range of the generator set based on the power regulation upper limit set of the generator set corresponding to each power transmission line; and determining an optimal power of each generator in the generator set in the output constraint range through a preset particle swarm algorithm. The technical scheme can reduce the iteration number of the particle swarm algorithm, reduce the time cost and computing resources consumed by searching, and improve the accuracy of the calculation result.

This invention discloses a method and system for expanding underwater image datasets based on a generative model. The method first constructs a pairing dataset of side-scan sonar images and text descriptions. After encoding the images into the latent space and adding noise, the LoRA module is used for fine-tuning while freezing the parameters of the basic diffusion model, enabling the model to learn the bright areas, shadow areas, and background texture features in the side-scan sonar images. Subsequently, new samples are generated based on preset prompts, and these generated samples are filtered and fused with the original samples to form an expanded dataset. This approach can improve the structural diversity and style consistency of side-scan sonar training samples, reduce data acquisition and annotation costs, and improve the training effect of downstream target detection models.

Embodiments of the present application provide a training method and device of a molecular optimal conformation prediction model. The method comprises: obtaining training data, wherein the training data comprises a molecular structure representation sample and an optimal conformation representation sample; processing the molecular structure representation sample by using a to-be-trained molecular optimal conformation prediction model to obtain a training output; constructing a loss function value according to the training output and the optimal conformation representation sample; and in a case where the loss function value or the number of training rounds is within a preset range, taking the trained to-be-trained molecular optimal conformation prediction model as the molecular optimal conformation prediction model. Embodiments of the present application can quickly and accurately predict the molecular optimal conformation without analyzing the entire conformation space of the molecule, thereby reducing the calculation resources and time consumed for analyzing the molecular optimal conformation.