490 results about "Sample sequence" patented technology

The invention discloses a flow field measurement method based on an event camera, and the method comprises the steps: generating a PIV data set, each time sequence sample sequence comprising a plurality of frames of continuous particle images, a corresponding velocity vector field, and particle event data at all moments; establishing a flow field data acquisition device based on an event camera and a high-speed camera, acquiring real event data and real image data which are synchronous in time so as to adjust parameters of an event simulator, and verifying and updating particle event data in the PIV data set according to the adjusted event simulator so as to obtain a flow field data acquisition result; obtaining the updated PIV data set as a training data set; building an event camera optical flow method model, and training by adopting the training data set; and on the basis of the trained event camera optical flow method model, event sequences in two adjacent time periods are used as inputs to calculate a velocity vector field corresponding to a middle moment. According to the invention, the flow field velocity field at the required moment can be obtained based on the event data within a period of time.

The invention discloses a network security malicious traffic tracing method based on a generative adversarial network, which comprises the following steps: collecting network traffic data, selecting a target traffic sample, extracting a protocol behavior, a communication structure and a time evolution characteristic of the target traffic sample, and carrying out semantic coding; mapping the multiple types of features to a unified embedding space through a multi-view fusion mode; constructing a manifold set based on known malicious samples, and generating an intermediate state sample sequence; constructing an attack context and path disturbance information in combination with the target sample, and generating a guide vector; inputting the guide vector and the random variable into a generator, and outputting a potential malicious sample representation; identifying the authenticity of the sample and evaluating the path similarity by using a discrimination network; and finally, constructing an attack path graph according to the similarity and the spatial relationship, and calculating a shortest path sequence to a known malicious sample to realize traceability identification of potential malicious behaviors. According to the invention, tracking analysis of malicious traffic sources is realized by using the generative adversarial network.

The invention provides a zero sample time sequence prediction method and system based on an adversarial neural network, and the method comprises the steps: constructing a time sequence generation module, training the generation module to stable convergence through obtained source domain data, and inputting collected random noise to generate a pseudo target sequence; constructing a time sequence prediction module, taking the pseudo sample sequence as a training set of the prediction module, and inputting the pseudo sample sequence into the prediction module for decomposition prediction; comparing the synthetic sequence with the output of the prediction module to obtain a prediction error, and feeding back the prediction error to the generation module to update generator parameters; through multiple rounds of generation-prediction-feedback closed-loop training, the obtained zero-sample linear architecture prediction model can effectively predict a corresponding future time sequence trend based on a synthetic pseudo sample under the condition that a target domain has no historical data completely. According to the method, data can be generated and self-training can be completed under the condition of no real data input, so that the quality of the generated data and the performance of the prediction model are synchronously improved.

The invention relates to the technical field of data processing, and discloses a secondary equipment hidden danger mining method and system based on a wave recording file and monitoring data. The method comprises the steps of calculating a conditional probability and intervention probability difference identification causal relationship by analyzing a recording file in combination with equipment topology and protection logic to construct a correlation graph, and performing Fourier transform and wavelet decomposition on a sampling sequence to extract multi-scale features; small signal test excitation is injected, a system identification estimation transfer function is fused with monitoring data to form a health feature vector to judge a health state, and a correlation map and the health state are input into a map neural network to calculate weighted attention coefficients and aggregate neighbor information to predict a fault propagation path; and establishing a degradation model, correcting the failure rate, predicting the residual life and generating a graded early warning report. According to the method and the device, the transformation from passive post analysis to active predictive maintenance is realized, and the timeliness, accuracy and systematicness of hidden danger identification of the secondary equipment are improved.

In the present disclosure a method and a system for generating an Optical Transport Network (OTN) DT network based on vertical federated learning are disclosed. The method includes: performing homomorphic encryption on a local fault root cause identifier to obtain an encrypted fault root cause identifier; receiving all encrypted alarm sample sequences corresponding to the encrypted fault root cause identifier; generating a single-domain training set according to the encrypted fault root cause identifier and the encrypted alarm sample sequences; training the local cross-domain fault root cause identification model according to the single-domain training set to obtain a model parameter update amount of the cross-domain fault root cause identification model; and reporting the model parameter update amount to the multi-domain orchestration system.

The invention relates to the technical field of urban rail transit station-city collaborative planning, in particular to an ontology-based station-city collaborative data integration and planning prediction method, which comprises the following steps of: obtaining rail transit station passenger flow data, resident travel behavior data and station periphery built environment index data; forming a space-time sample sequence according to the unified space-time granularity of the site walking service area; constructing an urban rail transit station-city cooperation ontology, and carrying out semantic annotation and semantic fusion on the space-time sample sequence to generate a feature sequence; inputting the feature sequence into a multi-task space-time diagram convolutional neural network prediction model to output a passenger flow prediction result and establish an environment index prediction result; and calculating a feature contribution degree based on a Shapley additive interpretation value, optimizing a background sample set by using a genetic algorithm to determine a key action element set, outputting a planning index threshold and an intervention measure parameter, and realizing an interpretable station-city collaborative prediction and planning decision closed loop.

The invention discloses a PCB component layout method, device and equipment based on a multi-modal large model, and relates to the technical field of electronic design automation, the method comprises the following steps: carrying out quality screening on a plurality of PCB design files which have been laid out to obtain a qualified design file set; constructing a serialized training sample set based on the qualified design file set; each training sample sequence comprises a plurality of training samples arranged according to a layout logic sequence; each training sample comprises multi-modal data representing a current layout state and a real position coordinate of a next component to be laid out; based on the serialized training sample set, training the open-source multi-modal large model to obtain a PCB component layout model; and iteratively generating a layout file of the target PCB based on the netlist file of the target PCB and the PCB component layout model. The method is high in generalization capability and low in cost based on the PCB component layout model trained by the layout data, and can realize rapid layout of components of a new PCB.

The embodiment of the invention provides a model training method, a carbon emission prediction method, a device and equipment. The method comprises the following steps: firstly, obtaining carbon emission sample sequence data; then, preprocessing the carbon emission sample data sequence to obtain preprocessed carbon emission sample sequence data; further, according to the preprocessed carbon emission analysis sample sequence data, determining a time sample feature and an interaction feature sample sequence; and then. Processing the sample time feature and the preprocessed power consumption sample sequence data to obtain a sample time feature component and a sample power consumption feature component; and finally, inputting the sample time characteristic component, the sample power consumption characteristic component, the interaction characteristic sample sequence and the preprocessed carbon emission sample sequence data into an initial Transform model for optimization training, and obtaining an improved target Transform model. In this way, the prediction precision of the prediction model and the generalization ability of the model are improved, and therefore accurate prediction of carbon emission data is achieved.

The invention requests to protect a multifunctional FFT (Fast Fourier Transform) hardware acceleration architecture capable of configuring software in real time, which comprises a first sample sequence rearrangement module, a first calculation unit, a second sample sequence rearrangement module, a second calculation unit, a third sample sequence rearrangement module, a butterfly operation circuit, a multiplexer and a dual-port RAM (Random Access Memory) which are sequentially connected end to end, in order to solve the technical problems that in an existing FFT hardware architecture, the operation length is fixed, the function is single, the fftshift function is lacked, and software dynamic configuration cannot be achieved, through software and hardware collaborative design, FFT hardware has operation flexibility, functional diversity and frequency spectrum processing capacity; the method can be widely applied to various digital signal processing scenes such as communication, images, radars and audios.

The invention discloses a distribution line early-stage fault time sequence hidden Markov modeling and identification method and system based on multistage evolution characteristics, and belongs to the field of distribution line early-stage fault identification. Comprising the steps of obtaining a current waveform sample sequence of an early fault in a distribution line; for each current waveform sample in the sequence, extracting a multi-dimensional time-frequency feature, and constructing a feature vector of each sample; performing fault stage identification on each sample by using the first-level hidden Markov model group, and outputting a fault stage tag sequence corresponding to each sample; combining the fault stage label sequences of the current sample and a plurality of previous historical samples to form a stage label sequence window; and respectively inputting the stage label sequence window into a second-level tree line fault hidden Markov model and a second-level non-tree line fault hidden Markov model, calculating a corresponding first average log-likelihood value and a corresponding second average log-likelihood value, comparing the two average log-likelihood values, and judging whether a current sample belongs to a tree line early fault or not.

The invention provides a cross-modal image-text retrieval method and system based on adaptive comparative learning. The method comprises the steps that two boundary parameters are introduced into probability distribution of each pair of positive samples, the probability of an image I corresponding to T is calculated, semantic consistency of each pair of positive samples is adjusted, and distribution characteristics of negative samples are distinguished and cloned; defining and calculating a saliency score of each batch of samples, and selecting negative sample sequences with the maximum saliency score and the minimum saliency score as an observation clone negative sample and a saliency negative sample; according to a Gaussian discriminant analysis method, predicting all potential clone negative samples of each batch of samples and generating anchor point samples; and iteratively updating the two boundary parameters according to the two boundary conditions of the similarity of the anchor point samples. According to the method, potential clone negative samples in each training batch are predicted by utilizing Gaussian discriminant analysis without additional training for clone negative samples easily appearing in an image-text data set. And a foundation is laid for subsequent further learning of potential high-dimensional key semantics of the cloned negative sample.

The invention relates to a rolling bearing residual life prediction method based on space-time degradation characteristic decoupling, and the method comprises the steps: obtaining a vibration signal in a full life cycle operation process of a rolling bearing, and obtaining a sample sequence after preprocessing; performing complete ensemble empirical mode decomposition on the sample sequence, and extracting a plurality of IMF signals; on the basis of statistical threshold criterion in combination with energy mutation and self-correlation structure mutation analysis, identifying a degradation starting moment, adding a label to a sample sequence, and dividing a training set and a test set; training the space-time degeneration decoupling network by using the training set to obtain an RUL prediction model; testing the RUL prediction model by using the test set to obtain a prediction result; the space-time degeneration decoupling network combines a degeneration guide feature deconstruction module and a collaborative modeling strategy of a time modeling branch and a space modeling branch, captures time dynamic characteristics and a space hierarchical structure, and improves the accuracy, stability and reliability of RUL prediction.

Methods, systems, and devices for wireless communications are described. A transmitter may modulate a set of bits into an on-off keying (OOK) sample sequence for wireless transmission to a receiver in a set of frequency resources. The transmitter may apply a transform (e.g., a discrete Fourier transform (DFT)) to the OOK sample sequence to generate a frequency domain representation of the OOK sample sequence. In some cases. the transmitter may, using an orthogonal frequency division multiplexing (OFDM) waveform generator, generate an OFDM waveform based on mapping the frequency domain representation of the OOK sample sequence to a set of resource elements of the set of frequency resources. In some cases. the transmitter may transmit the OFDM waveform to the receiver via the set of frequency resources.

The invention relates to the technical field of time sequence data processing and deep learning, in particular to a time sequence filling method and system based on a coarse-to-fine filling normal form. In the invention, a preprocessing module is used for sampling an input sequence to obtain a subsequence rk with the length of Lk, and a regression prediction module is combined with a causal mask to generate a prediction value r'k with the same structure and the same length as the rk based on all subsequences {r1... rk} generated by the preprocessing module; the correction module performs up-sampling on the predicted value r'k to obtain an up-sampling sequence r ''k with the length of T; supplementing missing data of the original sequence based on the r ''k to obtain a corrected sequence X (k + 1); and traversing k = 1... K to obtain a correction sequence X (K + 1) as a final repair completion time sequence. The method overcomes the defects that the time sequence filling mode in the prior art does not consider the unfixed missing rate and missing value block distribution, and is beneficial to improving the filling accuracy.

Provided is a training method for a fault detection model, a device fault detection method and related apparatuses, relating to the field of computer technology. The method comprises: sampling a designated device based on a preset drone formation to obtain a sample sequence; performing position encoding on the sample sequence according to the drone formation to obtain a drone formation encoding result; inputting the sample sequence and the drone formation encoding result into a model to be trained to obtain a fault detection result output by the model to be trained; determining a loss value based on the fault detection result and a true value of a fault detection result of the sample sequence; and adjusting a model parameter of the model to be trained based on the loss value to obtain the fault detection model.

The invention relates to the technical field of tumor data processing, in particular to a methylation-based tumor data processing system. The system comprises the following modules: a methylated sample sequencing module, a sequencing difference site recognition module, a tumor gene sequence analysis module and a tumor subtype classification module, performing high-throughput sequencing on the to-be-detected clinical tumor DNA sample to obtain methylated tumor sequencing data; performing methylation level quantification on the methylated tumor sequencing data, and performing tumor difference site analysis to generate tumor difference site data; performing tumor generation key gene sequence identification according to the tumor difference site data to generate tumor methylation characteristic data; and performing tumor subtype tag identification according to the tumor methylation characteristic data to generate tumor type tag data. According to the method, accurate tumor typing is realized on the basis of tumor DNA methylation characteristic analysis, and the false negative rate of low-concentration sample detection is effectively reduced.

The invention relates to a Parkinson's disease treatment effect prediction method based on a multi-modal image model and a related device. The method comprises the following steps: acquiring a multi-modal vector of a Parkinson's disease patient; aligning the multi-modal vectors on a time axis, and constructing a time point data element sample sequence; deploying a double-flow cross attention encoder for a sample sequence in each time point data element, and outputting a fused multi-modal feature through the double-flow cross attention encoder; and connecting the fused multi-modal features with digital clinical treatment scheme vectors corresponding to corresponding time points to form time point comprehensive feature vectors, inputting the time point comprehensive feature vectors to a time sequence information aggregation gating circulation unit, outputting final time point aggregation features, and inputting the final time point aggregation features to a multi-task adaptive prediction head. A UPDRS total score or a specific sub-scale score for the patient at a future preset point in time is predicted. According to the method, time sequence modeling is carried out on multi-mode and multi-time-point data, so that the accuracy and interpretability of Parkinson's disease treatment effect prediction are improved.

The invention relates to the technical field of artificial intelligence application, and discloses an enzyme EC number prediction method, and the method comprises the steps: obtaining the sample sequence characteristics of a to-be-predicted sample containing a substrate SMILES sequence and a product SMILES sequence through a target BERT model; constructing a molecular object and feature coding based on atom mapping, atom truncation and sequence analysis, constructing a reaction graph of a to-be-predicted sample, inputting the reaction graph into a target graph isomorphic neural network, and constructing molecular graph features of the to-be-predicted sample based on a recursive neighborhood aggregation mechanism; and fusing the sample sequence features of the to-be-predicted sample with the molecular map features by using a bidirectional cross attention mechanism to obtain multi-modal features, inputting the multi-modal features into the multi-layer perceptron, and obtaining the prediction probability of the enzyme EC number of the to-be-predicted sample. According to the method, efficient and accurate end-to-end prediction of enzyme EC numbering is realized through the multi-dimensional chemical spatial characteristics of the collaborative modeling reaction.

The invention relates to the technical field of anomaly prediction, in particular to a transformer operation anomaly prediction method and system, and the method comprises the following steps: obtaining the temperature, current, oil and vibration parameters of a transformer, aligning a sample sequence, calculating the Pearson correlation of the parameters, analyzing the mutation of the time sequence correlation, and extracting a feature group; and judging temperature difference and current variability, marking abnormal points, constructing a trend sequence to evaluate a risk trend, and outputting an abnormal trend prediction result. According to the method, by capturing linkage features between multiple parameter pairs in a time window, sudden change nodes can be identified, statistical features of multiple parameters in a time interval before and after an abnormal time point can be extracted, a composite feature set representing sudden change behaviors can be constructed, and an abnormal state can be judged under the combination condition of a maximum difference value and a variable coefficient. Risk evolution identification is realized by combining trending time sequence construction and abnormal frequency accumulation, and the stability and foresight of transformer operation abnormity prediction are improved.

The invention discloses a model training method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: inputting a plurality of state action pairs of a training sample sequence in a training data set into a decision model, and carrying out the decision processing; obtaining a target loss value corresponding to the decision model and a loss value corresponding to each training sample sequence; if the target loss value does not meet the preset convergence condition, updating model parameters of the decision model according to the target loss value; determining quality evaluation corresponding to the training sample sequence according to the loss value corresponding to the training sample sequence, and screening out the training sample sequence corresponding to the quality evaluation meeting a preset quality condition from the training data set to obtain a target training sample sequence; continuing to train the updated decision model according to the target training sample sequence; and if the target loss value meets a preset convergence condition, taking the decision model as a trained target decision model. According to the embodiment of the invention, the simulation learning effect of the decision model can be improved.

The disclosure provides a radar ranging method and device, a radar, and an in-vehicle system. An example method includes: obtaining a pulse waveform of a transmit signal sent to a target object by a radar; obtaining a sampling sequence of an echo signal received by the radar; determining, in sampling points on a rising edge of the sampling sequence, a first timing point used to indicate a receive moment of the echo signal; determining, based on the first timing point and the pulse waveform of the transmit signal, a second timing point used to indicate a transmit moment of the transmit signal; and calculating a distance between the radar and the target object based on the first timing point and the second timing point.

The invention provides a method for exploring performance goodness and generalization ability of various machine learning models in runoff sediment simulation. According to the method, the time sequence change trend of a river long-time sequence hydrological factor, a meteorological factor, a vegetation factor, an artificial activity factor and a runoff sediment transport change correlation factor is determined. Then identifying the correlation between each factor and the runoff sediment change, and selecting an important influence factor as an input factor of a machine learning model; the performance goodness of different machine learning models in simulation of runoff and sediment transport is evaluated, a sample data set is randomly divided to eliminate the influence of limited sample size and sample sequence on a simulation result, and each model is repeatedly trained and tested to determine the empirical risk, predictive ability and generalization ability of each model. And sorting the contribution degrees of the input factors in the runoff and sediment transport simulation process by applying a local interpretation method so as to reveal the main driving factors influencing the water and sediment change of the drainage basin.

The invention provides a medical resource utilization behavior assessment method and device, computer equipment and a storage medium. The method comprises the steps of obtaining a to-be-assessed behavior sequence; dividing the behavior sequence into a plurality of windows with the same time length, and converting each behavior point in each window into a corresponding behavior label; generating a sample sequence corresponding to the behavior sequence based on a plurality of windows of the behavior sequence and a plurality of behavior tags in each window; calculating an abnormal score of the behavior sequence based on the sample sequence; capturing a plurality of forward dependencies and a plurality of backward dependencies of the behavior sequence; splicing the plurality of forward dependencies and the plurality of backward dependencies to generate a feature vector; reconstructing the feature vector to obtain a reconstructed behavior sequence; calculating a reconstruction error value between the reconstruction behavior sequence and the behavior sequence; and judging whether the behavior sequence is credible or not based on the abnormal score and the reconstruction error value of the behavior sequence to be evaluated. According to the invention, whether the medical resource utilization behavior is credible can be accurately evaluated.

The invention provides a model training method for a transverse mixed attention mechanism, a medium, equipment and a program product, and the method comprises the steps: obtaining a data set containing a plurality of sample sequences, each sample sequence in the data set being formed by arranging a plurality of Token sequences obtained through word segmentation; constructing a to-be-trained model based on the pre-trained full attention model, and adding newly added parameters for linear attention calculation; in the same transverse mixed attention layer, executing total attention calculation on a Token set in a preset total attention calculation range, executing linear attention calculation on all Tokens, and fusing results of the total attention calculation and the linear attention calculation to obtain transverse mixed attention output used for forward reasoning and loss calculation; and based on the output and prediction result, only updating the newly added parameters to optimize the to-be-trained model until the to-be-trained model converges. According to the method, the calculation complexity and video memory occupation of long text sequence processing are reduced, and the reasoning speed and the resource utilization rate are improved.

The invention discloses a wind turbine generator generation power self-adaptive prediction method under non-stationary data flow, which comprises the following steps: S1, carrying out feature extraction on numerical weather forecast NWP data to obtain sample correlation features and sequence correlation features, and forming a sample-sequence feature vector; s2, similarity searching and clustering are carried out according to the sample-sequence feature vectors, and different meteorological modes are divided; s3, using the historical NWP and power data of each meteorological mode to train a Light GBM network offline to obtain a power prediction model, performing power prediction according to NWP meteorological data collected in real time, performing real-time monitoring on prediction performance of the power prediction model in each meteorological mode based on drift detection and a DDA algorithm, and adaptively adjusting model parameters to obtain a power prediction model; and adaptive prediction of the wind power under the non-stationary data flow is realized. According to the method, the change of the input and output mapping relation can be monitored in real time, the model is adaptively adjusted to adapt to a new mapping relation, and the power prediction precision is improved.

The invention relates to a training method of an amino acid sequence generation model and application thereof. The method comprises the following steps: acquiring training data, wherein the training data comprises sample structure information of sample protein; performing feature coding on the sample structure information to obtain node features and edge features corresponding to the sample protein; alternately updating the node features and the edge features through a plurality of message passing layers of a to-be-trained graph neural network model to obtain updated target node features and target edge features; according to the corresponding target node features and target edge features, decoding prediction is carried out on each node according to a random sequence through a to-be-trained graph neural network model, and prediction sequence information is output; and calculating network parameters of a loss value iteration graph neural network model according to the prediction sequence information and the sample sequence information, and obtaining a trained amino acid sequence generation model. According to the scheme provided by the invention, the generalization and robustness of the prediction result of the generated model can be improved.

The invention discloses an intelligent acquisition risk identification system based on deep learning, and the system comprises a time sequence sample construction module which is used for building a time sequence sample sequence; the hierarchical attention structural feature coding module is used for carrying out structural feature coding on the sequential sample sequence by utilizing a hierarchical attention network; the time sequence feature extraction module is used for inputting the structural feature vector into an ETSform model to perform time sequence feature extraction; the hierarchical time sequence collaborative attention adaptive fusion module is used for performing bidirectional attention interaction and dynamically generating a hierarchical weight and a time sequence weight through a meta-learning controller; the improved CatBoost risk identification module is used for outputting a risk score and a risk type label; and the system fusion module is used for summarizing and merging the risk scores and the risk type labels. The method and the device are suitable for merchant transaction risk identification in an acquiring business scene.

The invention relates to a training method of an RNA sequence modification prediction model, a prediction method and a related device. The method comprises the following steps: acquiring modification site sequencing data of an RNA sequence; on the basis of sites with target chemical modification and sites without target chemical modification in the modification site sequencing data, a plurality of positive sample sequences with different lengths and a plurality of negative sample sequences with different basic groups added at the central position are obtained respectively, so that a multi-level classification database is constructed; inputting each sample sequence in the multi-level classification database into a to-be-trained modification prediction model to obtain a prediction result of a modification site of each sample sequence; and on the basis of the prediction result and the actual condition of each sample sequence, updating model parameters of the to-be-trained modification prediction model until the trained modification prediction model is obtained. Compared with the prior art, the method can improve the training efficiency and precision of the model.

The invention discloses a self-adaptive virtual synchronous generator control method of an energy storage converter, and relates to the technical field of self-adaptive adjustment. The method comprises the following steps: S1, generating an electrical comprehensive state set based on three-phase voltage and three-phase current of a grid-connected point; s2, constructing a bearing trigger function and evaluating to generate a bearing trigger identifier; s3, obtaining a bearing fault diagnosis result based on the constructed diagnosis sample sequence; s4, according to the bearing fault diagnosis result and the diagnosis sample sequence, an adjustment parameter set is obtained, and an adjustment instruction is issued to the virtual synchronous generator; s5, updating the virtual power angle and the frequency reference; extracting a current vector and remodeling an original voltage reference according to an admittance parameter group; adjusting a power droop target and power angle evolution of a virtual synchronous generator of the energy storage converter according to a projection active instruction and a projection reactive instruction obtained through projection operation; the self-adaptive virtual synchronous generator control method realizes self-adaptive virtual synchronous generator control which keeps voltage shaping and virtual power angle continuity under current bearing constraint.

The invention discloses an RNA G-quadruplex prediction method and system based on a pre-training model and an RNA secondary structure, and the method comprises the steps: obtaining the position information of human rG4 on a human transcript; for each sequence, filling the same length to two sides according to the position coordinates of the sequence, so that the total length reaches a set length value; obtaining human cDNA sequence data as a reference sequence, and extracting rG4 data containing flanking sequence information from the cDNA sequence according to the filled sequence coordinates as a positive sample sequence; disorganizing each positive sample sequence to obtain a negative sample sequence; rNA secondary structure features of each sample sequence are generated; extracting sequence features of the sample sequence by using a pre-training model; and inputting the sequence features and the RNA secondary structure features into a prediction model for model training. According to the scheme, the secondary structural features of the RNA sequence are utilized, the secondary structural features are introduced as auxiliary information, and the prediction performance of the model can be remarkably improved.