35 results about "Gaussian mixture distribution" patented technology

The invention relates to the technical field of electric energy metering of a low-voltage power distribution system, and provides an EM algorithm-based district metering device misalignment identification method and system. The method comprises the following steps: establishing a theoretical misalignment value model of each outgoing line in different metering misalignment states, and determining a mapping relation between the different misalignment states and actual electric quantity data; the method comprises the following steps: constructing theoretical deviation values of each outgoing line metering device in different misalignment states on the basis of a transformer area incoming and outgoing line electric energy data and energy conservation relationship, estimating an initial attribution probability of a line state in combination with an index normalization mapping method, and obtaining a posterior attribution probability of each outgoing line line through Gaussian mixture distribution modeling under dynamic prior guidance; and based on the posterior affiliation probability of each outgoing line, constructing an objective function of joint observation value fitting quality, time continuity and misalignment amplitude balance, solving the objective function, and quantifying the electric quantity misalignment proportion of each outgoing line at each moment so as to judge classification and identification of misalignment behaviors.

This invention discloses a collaborative control method for highway merging zones based on deep reinforcement learning. A LiikeSim-Python co-simulation environment is established, and loop detectors are set up in the simulation environment to acquire traffic flow data upstream and downstream of the highway merging zone. An EM algorithm based on Gaussian mixture distribution is used as a traffic state classifier, taking the traffic flow data of the highway merging zone as input to classify the traffic state of the merging zone. A state space, action space, and reward function are designed. Using the state space of the highway merging zone as input and the actions of the variable speed limit agent and the ramp metering agent as output, a multi-agent shared experience network model under time-series characteristics is constructed. An independent experience pool is set up for the variable speed limit agent and the ramp metering agent, and the interaction experience between the agent and the traffic simulation environment is collected at the control cycle frequency. The agent model is trained using sampled data. The trained agent model is used to realize collaborative control of the highway merging zone. This invention can reduce travel delays in highway merging zones.

This invention discloses a fault diagnosis method for aero-engines based on an adaptive spatiotemporal decoupled graph convolutional network, comprising: acquiring and preprocessing multi-channel sensor signals to construct time-series samples; adaptively constructing an adjacency matrix to form a fault signal graph structure representing the spatial correlation of multi-source signals; inputting the fault signal graph structure into a graph convolutional network to extract spatial features; inputting the extracted spatial feature sequence into a bidirectional gated recurrent unit network to obtain temporal features; introducing a global attention mechanism to the temporal features to assign adaptive weights to features at different time steps; decoupling the spatiotemporal features using a variational autoencoder decoupling layer based on a Gaussian mixture distribution; and outputting the fault category of the aero-engine to complete the fault diagnosis. This invention significantly improves the accuracy, robustness, and engineering applicability of aero-engine fault diagnosis under complex operating conditions by jointly modeling the spatiotemporal characteristics of fault signals and effectively decoupling and fusing features.

The invention discloses a micro-earthquake dry and wet event separation method and device based on a Gaussian mixture model, and the method comprises the steps: S1, building a corresponding Gaussian mixture model based on a micro-earthquake detection result, carrying out the random initialization of parameters of the Gaussian mixture model, and setting the number of mixed components; s2, calculating the posterior probability of each microseismic event point relative to each Gaussian component; s3, updating Gaussian mixture distribution parameters based on the posterior probability of each microseismic event point relative to each Gaussian component; s4, based on the updated Gaussian mixture distribution parameters, calculating a maximum likelihood function and evaluating whether the maximum likelihood function accords with a threshold value, if not, returning to S2, and if yes, executing S5; and S5, outputting the Gaussian mixture model after parameter updating and a dry and wet event classification result. According to the method, effective transformation events corresponding to fracturing construction and fracture activity events caused by ground stress changes can be rapidly distinguished.

The embodiment of the application discloses a kind of reactive voltage control methods and devices of regional distribution network, the method comprises: obtaining each regional node sample corresponding to regional distribution network, each regional node sample includes the voltage and power factor corresponding to each regional node;The voltage and power factor corresponding to each regional node sample are input into Gaussian mixture distribution function, determine the target distribution probability corresponding to regional node sample;Based on target distribution probability and preset interval threshold, each regional node is clustered, to determine the target regional category cluster corresponding to each regional node;According to the electrical parameter characteristics corresponding to each target regional category cluster, determine the target control strategy of regional distribution network.Gaussian mixture distribution function is used for the clustering of regional distribution network VQC control, and the characteristics of each electrical parameter characteristic can be obtained according to the category cluster in the regional VQC control, and a targeted target control strategy can be used.

This invention relates to the field of wind power equipment fault diagnosis technology, and provides a method for extracting fault features from wind turbine gearboxes based on Gaussian mixture modeling. The method includes: modeling a mathematical model of the wind turbine gearbox observation signal by superimposing impact fault feature vectors and multi-source noise vectors, wherein the fault feature vector is the product of a redundant dictionary D and a sparse coefficient vector; modeling the multi-source noise vector as a Gaussian mixture distribution; constructing an objective function within a Bayesian framework to solve for the sparse coefficient vector in the mathematical model using maximum a posteriori probability estimation; simplifying the objective function; and using the EM algorithm and ADMM algorithm in a joint alternating iterative solution to obtain the optimized sparse coefficient vector; and reconstructing the impact fault feature vector in the wind turbine gearbox observation signal. This method improves the accuracy and robustness of extracting fault features from the observation signal of offshore wind turbine gearboxes.

The present application provides a kind of enterprise labor efficiency evaluation model construction method, medium and system, belong to artificial intelligence model technical field, the present application is by gathering staff behavior data to construct high-dimensional sparse feature matrix, after coding by hash embedding layer, gaussian mixture distribution modeling and double network cross screening noise correction are carried out to labeled label, employee collaboration relationship graph is constructed based on training sample set, after hierarchical random neighbor sampling and graph clustering presegmentation, employee performance evaluation vector is obtained by inputting causal inference enhancement double-flow comparative evaluation model, then the efficiency distribution is time series evolution by optimal transport gradient flow continuous flow algorithm, finally, according to the deviation of prediction result and equilibrium threshold, external incentive term is adjusted and intervention strategy is output, the technical problem that employee performance evaluation model cannot accurately estimate causal effect in organization network and realize group efficiency time series evolution prediction is solved.

The application discloses an industrial process fault detection method and system based on double subspace division, which firstly performs normality evaluation on process variables, divides a data space into Gaussian subspace and non-Gaussian subspace; a fault detection model based on principal component analysis is established in the Gaussian subspace, and a fault detection model based on support vector data description is established in the non-Gaussian subspace; current test data variables are divided into the Gaussian subspace and the non-Gaussian subspace, and are respectively input into the trained models, so that fault detection statistics corresponding to each subspace are calculated; each subspace statistic is converted into a fault probability by using Bayesian inference and is weightedly fused to construct a comprehensive monitoring statistic, and fault detection is realized. The application effectively solves the problem that a traditional method has poor adaptability to Gaussian and non-Gaussian mixed distribution data, improves the fault detection rate while significantly reducing the false alarm rate, and is suitable for real-time monitoring requirements of complex industrial processes.

The application provides a target detection method, device and equipment based on a Gaussian mixture model, which can be applied to the field of computer vision and the field of digital image processing. The target detection method based on the Gaussian mixture model comprises: processing a target image comprising a target to be detected to obtain a first density map related to the target to be detected, wherein the first density map conforms to a Gaussian mixture distribution, and the first density map serves as a Gaussian mixture model; performing integral processing on the first density map to obtain a first quantity value of the target to be detected; performing local peak value extraction on the first density map to obtain a second quantity value of the target to be detected; obtaining a target mean value and a target variance of the target to be detected based on the first quantity value, the second quantity value and the first density map; and determining position information of the target to be detected according to the target mean value and the target variance.

The application discloses a soft measurement modeling method based on mode perception dynamic variational autoencoding regression, and comprises the following steps: (1) acquiring three-phase flow process data with multiple mode dynamic characteristics; (2) data division and pretreatment operation; (3) establishing a mode perception dynamic variational autoencoding regression model, and realizing the prediction of an online quality variable; and (4) predicting a three-phase flow process pressure variable and performing model performance evaluation. The encoder-decoder network with a convolution-deconvolution structure learns the hidden layer feature representation of dynamic data, the multiple mode characteristics of the data are mined in the hidden space by using a Gaussian mixture distribution, the mapping relationship between the latent variable and the key quality variable is effectively learned, in addition, the reconstruction of dynamic multiple mode data is constrained by using a Wasserstein distance, and finally the purpose of improving the accuracy of key quality inference is achieved.

The present application belongs to the technical field of power system operation, and relates to a double-sided random power grid scheduling method. By analyzing historical data of wind power output, Gaussian mixture distribution fitting is carried out by using software. For determined power system parameters, a double-sided chance constraint random scheduling model is established. Then, the hyperbolic tangent function is used to analytically approximate the cumulative distribution function of random variables in the reserve demand constraint and the line flow constraint, so as to convert the double-sided chance constraint into a deterministic constraint, and convert the original problem into a convex optimization problem which is easy to solve. Finally, the scheduling model is solved to obtain the scheduling strategy. The present application has the advantage that the double-sided chance constraint containing the risk level and the random variable is converted into a solvable deterministic convex constraint by using the hyperbolic tangent function, which effectively improves the solving efficiency of the model and provides a more reasonable scheduling basis for decision makers. The present application can be applied to active power scheduling of a power system containing large-scale renewable energy grid-connected.

A residual life prediction method based on domain invariance and consistent ordinal number representation learning comprises the following steps: firstly, calculating an MMD distance between different bearing domains in a training set, and minimizing the distance through gradient descent to enable feature distribution of each source domain to be globally aligned, extracting domain-invariant and ordered degeneration features, introducing a Gaussian mixture prior coding-decoding structure, and finally obtaining a residual life prediction result; the features are mapped to a potential space constrained by Gaussian mixture distribution through a variational auto-encoder, and the inadaptability of traditional single Gaussian distribution to nonlinear degradation is reduced; and in combination with MMD alignment, ordinal number loss and local consistency constraints, the dominant situation of a single constraint is avoided, the feature space has domain invariance and continuous orderliness at the same time through joint optimization, and the generalization ability of the model in a cross-domain task is remarkably improved.

The present application relates to a kind of photovoltaic module unsupervised defect detection methods based on contrast learning, including using normal sample training contrast learning network model;Contrast learning network model contains two encoder networks and a object-oriented symmetric cross attention network;Based on encoder network, the feature distribution of normal sample is obtained, and the defect discrimination model for normal sample feature is constructed using multivariate Gaussian mixture distribution;The image to be tested is input into the encoder network to obtain features, and the detection result of image is obtained based on defect discrimination model.The beneficial effects of the present application are: the present application can effectively detect the tiny and diverse defects of photovoltaic module, only normal photovoltaic image is needed for training, and the problem of sample imbalance can be solved;And in the case of few defect images, fast and accurate detection of unknown defects can be realized, with the advantages of environmental adaptability and strong robustness.

The invention discloses a picture style migration method, and belongs to the technical field of image style migration, the picture style migration method comprises the following steps: S100: inputting a picture; s200, using noise to destroy the input picture, adding Gaussian mixture distribution processing based on diffusion in the destroying process, and using the processed picture as a sample; s300, inputting the sample into the style model, and converting to obtain a new image; and S400, denoising the new image, and outputting the new image to complete image style migration. According to the invention, picture style migration can be effectively, quickly and accurately realized.

The invention discloses an ore grinding particle size robust soft measurement method based on double Gaussian mixture distribution. The method comprises the following steps: constructing a random configuration network as a basic prediction model; secondly, explicitly establishing a noise model weighted and mixed by a small variance Gaussian component and a large variance Gaussian component, and respectively fitting conventional measurement noise and process abnormal noise; under a Bayesian framework, an expectation maximization algorithm is adopted to jointly iteratively optimize network output weight and noise model parameters, and adaptive learning of the parameters and intelligent weighting of samples are achieved. According to the method, noise decomposition parameters including the normal working condition weight and the abnormal variance ratio can be output, and interpretable decision support is provided for ore grinding operation. Experiments show that the method effectively improves the prediction precision, robustness and interpretability of the model in the mixed noise environment, and provides reliable technical support for optimization control of the ore grinding process.

The invention discloses a charging pile remote metering method and system based on multi-source data fusion, and belongs to the technical field of safety and credibility verification of charging pile metering data. Collecting charging pile electric energy data, user vehicle BMS data and electric energy data of a power grid power supply side; uploading based on hybrid communication and a time synchronization mechanism to realize uploading of multi-source data carrying a unified time label; constructing a Gaussian mixture model error distribution model to obtain Gaussian mixture distribution; based on a Bayesian formula, calculating a charging data credibility probability under the condition that data of a pile end, a vehicle end and a charging station level power grid end are given; calculating the credible probability of the charging data; and performing data fusion and correction according to the obtained credible probability of the charging data to obtain corrected charging data. The safety of the metering data of the charging pile is improved, and tampering and fraudulent behaviors are prevented.

This invention belongs to the field of data amplification technology, and relates to a method, system, and terminal for amplifying electroencephalogram (EEG) signals based on a Gaussian mixture model. The amplification method includes: acquiring and obtaining raw EEG signals including lead signals; obtaining multiple extreme points for each lead signal and segmenting the lead signal; constructing a phase-shifted Gaussian mixture model for each extreme point for each lead signal, and randomly sampling the phase difference from the phase-shifted Gaussian mixture model corresponding to the extreme point; applying the phase difference to the corresponding extreme point, adjusting the sampling point using upsampling / undersampling to achieve peak shift, obtaining the initial amplified signal corresponding to each lead signal, and smoothing and filtering it to obtain the filtered amplified signal; constructing an amplitude-scaling Gaussian mixture model for the raw EEG signal, and sampling amplitude variation parameters from it; applying the corresponding amplitude variation parameters to the filtered amplified signal for each lead signal to generate amplitude variation, ultimately obtaining the amplified signal.

The invention discloses a horizontal and vertical seismic oscillation combined selection method based on a multivariable return period, and the method comprises the steps: determining the annual exceeding probability of a site, namely a target return period, and selecting a condition strength parameter and a target horizontal and vertical strength parameter; according to the basic information of the site, obtaining a random seismic directory of the site through Open Quake, and determining the occurrence rate of an earthquake year; constructing multivariate Gaussian mixture distribution based on a site random seismic directory; randomly extracting a large number of simulation spectrums from the multivariate Gaussian mixture distribution; calculating a multivariable recurrence period of each simulation spectrum, and selecting the simulation spectrum with the recurrence period within + / -5% of the target recurrence period as a target spectrum; after the target spectrum distribution is known, a seismic oscillation record set meeting the error requirement is selected from the target database; different from a single-variable return period based on a single-dimensional strength parameter, a horizontal and vertical seismic oscillation selection method based on a multivariable return period can select a group of horizontal and vertical seismic oscillation records with danger consistency.

This invention proposes a federated learning method robust to mixed noise, comprising: sending local metric model parameters updated by the client based on local training data and a subjective logistic loss function to the server; obtaining global metric model parameters calculated by the server based on the local metric model parameters updated by the client and other clients, and the corresponding sample data volume; the client calculating the subjective logistic loss and local Gaussian mixture distribution for each training sample based on the global metric model parameters and local training data, and sending them to the server; obtaining interval thresholds calculated by the server based on multiple local Gaussian mixture distributions, and performing mixed noise identification on the training data based on the interval thresholds and local Gaussian mixture distributions, thereby filtering and correcting the identified open set noise and closed set noise respectively. This invention also proposes a federated learning system robust to mixed noise, and a data processing device for federated learning.

This invention discloses a few-sample image generation method for image generation in scenarios with limited data. The few-sample image generation method provided by this invention includes: randomly sampling from a dynamic Gaussian mixture distribution to obtain a dynamic Gaussian mixture latent code; inputting the dynamic Gaussian mixture latent code into a generator network, enhancing the intermediate features of the generator network through a hybrid attention mechanism, wherein the intermediate features are obtained by mapping the dynamic Gaussian mixture latent code by the generator network, and inputting the enhanced intermediate features into the generator network to obtain a generated image set; inputting the generated image set and a real image set into a discriminator network to obtain image discrimination results for the generated image set and the real image set; updating the generator network and the discriminator network according to the image discrimination results and the objective optimization functions of the generator network and the discriminator network to obtain updated generator networks and discriminator networks.

The application discloses a subway network passenger travel space-time path estimation method based on fusion incremental learning. The method obtains passenger entry and exit card swiping records through an AFC system of a subway network, combines with ATS train operation data, constructs a space-time network and generates a passenger feasible space-time path set. Taking a passenger of a single feasible space-time path as a sample basis, a Gaussian mixture distribution model is used to model a walking time distribution. For a multi-space-time path passenger, the posterior probability of each space-time path is calculated by combining with a Bayesian theorem, and dynamic estimation of a passenger path selection probability is realized. Through an incremental learning mechanism, model parameters are adaptively adjusted according to the confidence of the space-time path, and the model is updated in real time to adapt to changes in passenger behavior patterns. The application solves the problems of small sample size, poor scalability and high computational complexity in the prior art, can efficiently and dynamically estimate passenger space-time paths, and provides decision support for subway operation management.

The present application relates to the field of sensor information fusion, in particular to a probability distribution mean sensor fusion method based on limited mixed model optimization. Any probability density can be approximately expressed in the form of Gaussian mixture, the method aims to express the posterior distribution of the target state in the form of Gaussian mixture, when performing sensor fusion, the patent solves the KL divergence between Gaussian mixture distribution through the variational method, finds a best fusion weight by minimizing the KL divergence of AA fusion to the real target state, so that the result of sensor fusion is closer to the real target state, and the target is more convenient to estimate and track.

This invention discloses a method for predicting count-type quality variables based on a variational Bayesian Gaussian-Poisson mixture regression model. This method can be used for data analysis and prediction where the dependent variable is count data and the independent variable is continuous numerical data. Its core involves fitting continuous and count data respectively using Gaussian mixture distributions and Poisson mixture regression distributions, assuming that the two mixture distributions share the same mixing coefficients, and employing variational inference techniques for model parameter learning. This invention overcomes the limitation of traditional soft sensing methods in providing discrete probability estimates for count data and can solve the problem of process variables and quality variables exhibiting multiple modalities due to multiple operating conditions in industrial processes.

The application discloses a kind of probability Gaussian embedding knowledge tracking methods based on knowledge relationship perception, comprising the following steps: modeling the relationship between questions, the relationship between questions is inferred by the record of the history of students in different questions answer.Statistical question-knowledge concept hybrid graph is constructed, and knowledge concept is mapped to probability distribution to model the hierarchical containing relationship between different knowledge concepts. The hierarchical containing relationship between knowledge concepts is expressed by modeling them as Gaussian distribution, so that each knowledge concept is regarded as a Gaussian distribution, and the mean value represents the central position of the concept, and the standard deviation reflects the range of the concept.To represent the partial mapping relationship between questions and knowledge concepts, they are first mapped to the same representation space, and then a Gaussian mixture distribution is used to achieve the goal.The overall knowledge state of students is modeled by their historical answer records, so as to infer the answering situation of students at the next moment.The application is obviously accurate in predicting the answering situation of students, and can provide more accurate and personalized learning content recommendation for students.

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive a configuration associated with an interference report. The UE may transmit the interference report in accordance with the configuration, wherein the interference report includes an indication of one or more parameters of a Gaussian mixture distribution that is fitted to an interference information distribution. Numerous other aspects are described.

The invention discloses an enhanced modeling and state estimation method for a turboshaft engine oriented to unknown noise, and aims to solve the problems of low modeling precision and large estimation error of a traditional model under health parameter degradation and unknown noise. According to the method, a state space model is established through a small perturbation method, then an enhanced linear model containing state and output compensation is constructed, and the prediction precision of an airborne model is improved; meanwhile, designing a noise acquisition module based on a Gaussian mixture model (GMM), converting unknown noise into Gaussian mixture distribution, and obtaining statistical characteristics; and finally, combining an enhanced linear model, a GMM (Gaussian Mixture Model) module and linear Kalman filtering to build a state estimation system. According to the method, high estimation precision can still be kept in a multi-health parameter degradation and unknown noise scene, the airborne real-time requirement is met, and the method is of great significance to engine fault diagnosis and health management.

This invention proposes a method and system for extracting T-wave morphological features using a Gaussian mixture variational autoencoder, belonging to the field of electrocardiogram (ECG) signal processing technology. It addresses the problems of limited morphological description, fixed-length input limitations, and restrictive distribution settings in existing ECG T-wave analysis. The method constructs a labeled T-wave dataset; designs a temporal-morphological attention encoder to map variable-length sequences to fixed-dimensional latent variables; constructs a Gaussian mixture variational autoencoder to force features in the latent space to follow a multimodal Gaussian mixture distribution, enabling automatic clustering of T-wave morphologies; and introduces a T-wave curvature perception loss during the decoding and reconstruction stage to accurately restore the sharpness and smoothness of the T-wave. This invention solves the problems of existing technologies, achieving accurate extraction and classification of morphological features.

The application relates to a kind of component extreme effect short-time evaluation methods based on bridge deck vehicle distribution, comprising: collecting and statistics the vehicle flow and vehicle data information of specific bridge site, obtaining the influence line of bridge structure to be evaluated effect;Vehicle flow data is loaded to the influence line of to be evaluated effect, and extreme value scene sample is obtained;The Poisson parameter of each lane heavy vehicle distribution position and the Gaussian mixture distribution parameter of total weight of heavy vehicle are counted;Based on non-stationary Poisson distribution, heavy vehicle position simulation is carried out, and based on Nataf transformation, total weight of heavy vehicle is simulated, to obtain load simulation extreme scene under to be evaluated effect;Using load simulation extreme scene under different effects, the complex spatial distribution of heavy vehicle on bridge deck is simulated.Compared with the prior art, the application can quickly and accurately simulate the heavy vehicle distribution on the bridge deck that causes the extreme value of various effects of the bridge, thereby ensuring the efficiency and reliability of the bridge component extreme effect evaluation.

The invention relates to an electric vehicle adjustable capacity dynamic estimation method fusing multi-factor interaction and non-complete rational decision, and belongs to the field of electric power systems. The method comprises the following steps: simulating the travel time and path of an electric vehicle based on a travel chain theory and Gaussian mixture distribution, and predicting a charging demand in combination with a dynamic energy consumption model; establishing a vehicle-road-station multi-factor information interaction model, fusing a traffic road network state, charging station queuing information and a user remaining endurance mileage, and realizing accurate simulation and adjustable capability evaluation of a vehicle access behavior of an electric vehicle network access station; constructing a non-complete rational user behavior model based on reference point dependence, comprehensively considering energy consumption cost saving income and off-network time delay loss, and solving an optimal charging and discharging strategy of a user; and finally, the charging and discharging adjustable capacity of the cluster electric vehicles in the space-time dimension is evaluated. According to the method, the interaction influence of the user decision behavior and the system can be reflected more truly, and the accuracy and practicability of adjustable capacity evaluation are improved.

The invention provides a probabilistic wind speed set prediction method and system based on short-term historical data driving, and relates to the field of meteorological prediction.The method comprises the steps that a wind speed prediction model is constructed and trained, and the wind speed prediction model comprises a deterministic encoder unit, a distribution predictor unit, a set sampling unit and a physical consistency constraint unit; the deterministic encoder unit is used for extracting feature vectors from an input historical wind speed sequence, the distribution predictor unit is used for predicting Gaussian mixture distribution parameters of a plurality of future time steps based on the feature vectors, and the set sampling unit is used for generating a plurality of future wind speed trajectories based on the Gaussian mixture distribution parameters of the plurality of future time steps. The physical consistency constraint unit is used for applying physical consistency constraint to a plurality of future wind speed trajectories to generate a wind speed forecast set; through the trained wind speed prediction model, the wind speed prediction set is generated based on the historical wind speed sequence of the current window, and the method has the advantage of improving the accuracy of wind speed prediction.