319 results about "Probability vector" patented technology

The invention discloses an intelligent rock weathering detection method based on a multispectral imaging technology, and particularly relates to the field of rock detection.The method comprises the steps that a multispectral imaging device is used for collecting a rock sample image, and then radiometric calibration, geometric correction and image enhancement processing are sequentially conducted on the image. The method comprises the following steps of: preprocessing an image, acquiring spectrum, texture and morphological characteristic parameters from the preprocessed image, collecting samples through a multi-source database, constructing a deep convolutional neural network model, inputting the acquired characteristic parameters into the model to obtain a rock weathering probability vector, judging the weathering degree according to a maximum probability criterion, and outputting a result to a user side. According to the method, multispectral imaging and deep learning technologies are fused, the defects of a traditional detection method are overcome, the rock weathering degree can be detected more accurately and efficiently, nondestructive detection and real-time monitoring are achieved, and the method has important application value in the fields of geological research, engineering construction and the like.

The invention discloses a weld defect detection method and system, and relates to the technical field of defect detection.The method comprises the steps that a visible light image sequence, an infrared image sequence and a pixel point cloud sequence are collected, windows are moved in geometric coding results of pixel point cloud of each frame and a visible light image through window attention, and the windows are extracted and sorted into visual feature maps; extracting a temperature feature map from each frame of infrared image through thermal gradient convolution and cavity convolution, and arranging the temperature feature map into a geometric feature sequence and a defect feature sequence based on cross attention and decoupling head mapping; capturing the time sequence dependence of the defect feature sequence through a Transform encoder, carrying out potential space modeling, and generating a defect probability vector by utilizing defect priori knowledge in combination with a Bayesian network; and a Gaussian regression model and a logarithmic probability equation are adopted to map the geometric feature sequence into a probability threshold of each defect type, and the probability threshold is compared with a defect probability vector to determine the defect, so that high-precision defect detection fusing geometry, temperature, time sequence and priori knowledge is realized.

The invention discloses a fault diagnosis method and device, computer equipment and a computer readable storage medium, relates to the technical field of power systems and automation thereof, and solves the problem of fault misjudgment caused by important fault characterization of a method which is easy to lose when single modal information is used for fault diagnosis at present. The method comprises the following steps: converting a bus voltage signal into a two-dimensional time-frequency diagram based on a continuous wavelet transform model, and converting the bus voltage signal into a one-dimensional frequency spectrum sequence based on a fast Fourier transform model; performing feature extraction on the two-dimensional time-frequency graph based on a time-frequency image feature extraction model to obtain a first feature vector, and performing feature extraction on the one-dimensional frequency spectrum sequence based on a frequency spectrum sequence feature extraction model to obtain a second feature vector; performing feature fusion on the first feature vector and the second feature vector based on a feature fusion model, and obtaining a category probability vector by using a fault diagnosis model based on the fused feature vector; and determining the highest probability value in the category probability vector, and taking the corresponding fault type as a target fault type.

The invention relates to the technical field of intelligent decision making, in particular to a multi-source reasoning-based confrontation environment target behavior analysis method and system. Comprising the following steps: detecting a blue target under a simulation system, obtaining target position and state information, and generating a knowledge graph; dynamically extracting an entity sub-graph associated with the blue target through a space-time constraint graph sampler; constructing a dual-channel decision generator to respectively output an action probability vector driven by an expert rule and an action probability vector derived by a Bayesian network; designing a high-dimensional decision space projection module to map the two types of probability vectors into n-dimensional space coordinate points, and performing probability trajectory fusion and error correction through an improved Kalman filter; and a game utility corrector is introduced to optimize the final action decision based on the Nash equilibrium principle. The problem that in the prior art, a static knowledge graph is difficult to support real-time fusion of multi-source decision information in a dynamic confrontation environment is solved, and the timeliness and accuracy of complex situation behavior reasoning are remarkably improved.

The invention discloses a long-tail SKU sales prediction method and system, and belongs to the technical field of sales prediction, and the method comprises the following steps: obtaining SKU sales feature data, and carrying out the preprocessing of the sales feature data; obtaining a multi-dimensional feature table based on the standardized training data set; a rule engine and a lightweight classification model are used for joint judgment to classify the multi-dimensional feature table, and four SKU category labels and probability vectors thereof are output; differentiated training models of the four SKU category labels are established, different types of SKUs are routed to the corresponding training models, and SKU sales volume point predicted values are output; probability calibration is carried out, and a plurality of quantiles are calculated to form interval prediction; establishing an SKU multi-hierarchy relationship, and performing hierarchy consistency solution on prediction results among multiple hierarchies; in the online operation process of the prediction model, input data distribution drift is monitored, and model updating is triggered when the input data distribution drift exceeds a threshold value. According to the method, the prediction precision and stability of the long-tail SKU are improved, and rapid adaptive prediction of sudden and off-peak demands is realized.

The invention relates to the technical field of on-line monitoring, in particular to a flue gas cooler leakage monitoring system and method.The method comprises the steps that a multi-source real-time sensing layer module continuously collects tube plate weld joint area temperature data, cooling water flow characteristics and flue gas sulfur-containing gas concentration, and the sampling period is 250 milliseconds; the intelligent feature fusion module performs noise reduction processing and dynamic weight fusion on the multi-source data and outputs a leakage probability vector; the dual-system collaborative early warning module maps the probability vectors into confidence levels, a 65% confidence threshold triggers an acoustic verification system to start, an acoustic emission sensor array captures fractured sound wave features and then compares the fractured sound wave features with a voiceprint library through a convolutional neural network, and an early warning signal is output when the fractured sound wave features exceed a 0.85 similarity threshold; and the dynamic knowledge base updates the case characteristics online according to the early warning result. According to the system, second-level identification and continuously optimized closed-loop monitoring at the initial stage of leakage germination are realized, and the problem of response delay caused by intermittent detection in the prior art is solved.

The invention discloses a micro-seismic event identification method, device and equipment and a storage medium, the disclosed micro-seismic event identification method is applied to a graph neural network, and the graph neural network is composed of a plurality of mutually connected nodes. According to the application, the response result of each sensor node is obtained by responding to the stress signal of the signal node through each sensor node, and the response result at least comprises waveform data and audio data; inputting each waveform data into a preset waveform prediction model to obtain an original waveform probability vector corresponding to each sensor node; inputting each piece of audio data into a preset audio prediction model to obtain an original audio probability vector corresponding to each sensor node; and according to each original waveform probability vector and each original audio probability vector, determining the probability that the stress signal is triggered by the preset micro-seismic event, thereby realizing automatic identification of the micro-seismic event without depending on artificial experience, effectively improving the identification speed of the micro-seismic event, and further improving the timeliness of micro-seismic early warning.

The invention belongs to the technical field of spectral analysis, and discloses a Raman spectrum characteristic peak segmentation method based on distance vector and probability vector output. Spectral detail features are extracted through multi-scale wavelet decomposition, and a spectral multi-layer structure representation matrix is constructed; identifying a potential peak site and a hierarchical affiliation relationship thereof based on the local curvature change rate; generating an adaptive distance calculation kernel function in combination with the asymmetry index and the peak shape complexity coefficient; and calculating a distance vector of hierarchical perception and a probability vector extracted by deep learning, and constructing a joint segmentation decision function of a multi-layer peak structure. According to the method, the hierarchical relationship among the main peak, the shoulder peak and the sub-peak can be accurately distinguished, the complex conditions of peak overlapping, asymmetric peak shapes, low signal-to-noise ratio and the like are effectively processed, and the accuracy and the reliability of Raman spectrum analysis are improved.

The invention provides an electro-permanent magnet uncoiling composite magnetic field multi-mode adjustment system which comprises a data acquisition and preprocessing module, a graph structure construction module, a dynamic prediction module, a historical matching prediction module, a magnetic force adjustment strategy module and an electro-permanent magnet execution mechanism. The data acquisition module constructs a multi-modal data cube through a temperature field, deformation, stress and magnesium powder concentration sensing network; the graph structure module discretizes the steel strip coil into a node network and calculates a mechanical association edge weight; the dynamic prediction module updates node features through the graph neural network and then outputs a sliding probability vector through the LSTM network; the historical matching module screens similar working conditions based on a dynamic time warping algorithm and generates magnetic force preset parameters; the strategy model fuses the probability vector and a preset parameter to generate a partition magnetic matrix; and the execution mechanism realizes millisecond-level magnetic field regulation and control through the reconfigurable magnetic pole array. According to the system, the magnetic field control precision can be improved, the interlayer sliding risk is reduced, and the self-adaptive adjustment effect is achieved.

A method for multi-label image classification in a convolutional neural network (CNN) is provided that includes forming a composite image from a plurality of clipped images, and processing the composite image by the CNN to generate a probability vector for each clipped image of the plurality of clipped images, wherein a length of a probability vector is equal to a number of classes the CNN is designed to classify.

The invention discloses a fault fusion diagnosis method of a distributed photovoltaic system and a related device, and belongs to the technical field of fault diagnosis, and the method comprises the steps: collecting sensor time sequence data and an infrared image of the photovoltaic system, and carrying out the preprocessing; extracting an original time sequence feature and a frequency domain feature of the sensor time sequence data, inputting the original time sequence feature and the frequency domain feature into a pre-trained SVM model to obtain a second probability vector, and inputting the infrared image into a pre-trained CNN model to obtain a first probability vector; performing weighted fusion on the first probability vector and the second probability vector to obtain a weighted score vector; and based on a preset threshold, performing confidence calculation on the weighted score vector, and after a confidence condition is satisfied, obtaining a fault category, and completing fault fusion diagnosis. According to the invention, the problem of misjudgment or missed judgment easily caused by direct output of a single diagnosis result during fault diagnosis in the prior art can be solved.

The invention discloses a gate fault detection method and system, and relates to the technical field of fault diagnosis and health monitoring, and the method comprises the steps: collecting a gate vibration signal, extracting a main modal frequency, setting a sampling frequency and a passband range, designing a finite impulse response filter for filtering, and carrying out the optimization through a particle swarm algorithm, thereby obtaining an enhanced signal; extracting features in the enhanced signal for integration to obtain a feature vector, mapping the feature vector to a discrete state set by using a rule threshold matching method, and obtaining a gate state; the method comprises the following steps: acquiring historical data for sampling, constructing a state transition probability matrix, calculating a steady-state probability vector according to the state transition probability matrix, combining a gate state, generating a state sequence track, setting an influence factor and a weight by using an empirical rule, and calculating a fault occurrence probability for risk detection; according to the method, the gate fault detection precision, efficiency and reliability are remarkably improved.

The invention relates to a cross-modal association reasoning method, device and system based on small sample learning, and relates to the technical field. The method comprises the following steps: acquiring multi-source heterogeneous data, pre-processing the multi-source heterogeneous data, calculating respective characteristic values of the pre-processed multi-source heterogeneous data by utilizing a characteristic vector representation model, and inputting the respective characteristic values of the multi-source heterogeneous data into a trained association reasoning model, the association reasoning model is obtained by performing association reasoning based on respective characteristic values of the multi-source heterogeneous data, the association reasoning model is obtained by training a meta-learning model by adopting a meta-learning method, and an association reasoning result of the multi-source heterogeneous data is determined by utilizing the association probability vector. According to the scheme, cross-modal reasoning of heterogeneous data can be realized in a small sample data environment.

The invention discloses a drainage basin unit debris flow susceptibility evaluation method based on multi-model Stacking integration. The method comprises the following steps: constructing a debris flow disaster data set; constructing a Stacking ensemble learning model which comprises four base models, a feature stacking layer, an attention layer, a mixed feature layer, a feature splicing layer and a meta-model; a Stacking integrated learning model is trained; predicting the debris flow disaster probability of the drainage basin unit to be identified; and performing debris flow susceptibility evaluation. According to the method, multi-source data can be efficiently fused, and the mixed probability vector and the original sample are creatively spliced, so that the meta-model can simultaneously utilize the intrinsic characteristics of the original data and the prediction information of the base model in the training process; interference of irrelevant information and dependence on the quality of a base model can be reduced, key information in data can be captured more comprehensively, and the explanation capacity and prediction precision of the model on complex data are enhanced.

A data processing method includes obtaining N sample image sets and an image recognition model, obtaining a sampled sample image from the N sample image sets, inputting the sampled sample image and the wrong class label into the image recognition model to generate a first probability vector of the sampled sample image for the N class labels, obtaining, from the N probability elements in the first probability vector, a first probability element indicating the wrong class label, and adjusting the sampled sample image based on the first probability element, to obtain an adversarial sample image corresponding to the sampled sample image.

The invention provides an instruction output method and device based on a brain-computer interface, a medium and electronic equipment. The instruction output method based on the brain-computer interface comprises the steps that electroencephalogram signals of a tested object are obtained based on the brain-computer interface; at each refresh point, encoding the electroencephalogram signal within a fixed time length, and generating an encoding probability vector of the electroencephalogram signal at the refresh point; based on the coding probability vector, decoding and generating an instruction output probability vector; and determining discretely output preset instructions and the number of the instructions based on the execution probability of each preset instruction in the instruction output probability vector. According to the method, the continuously input coding probability vector is analyzed in real time, so that the change of the electroencephalogram signal can be quickly responded, and the execution probability of the instruction is updated in time. According to the method, a single instruction can be output, the output of a plurality of instructions can be determined according to the execution probability, the complex user intention is more effectively responded, and the practicability of the electroencephalogram control system in various application scenes is expanded.

The invention discloses a context poisoning attack defense method oriented to an RAG application, and relates to the technical field of RAG. the method comprises the following steps: inputting a target query statement, and retrieving the target query statement to obtain multiple pieces of context information; taking representative sentences in the retrieved context information, and identifying and filtering potential malicious template clusters; the big language model gives all candidate answers according to existing context information, the logarithmic probability of all contexts to different candidate answers is calculated, and after the influence of parameter knowledge of the big language model is removed from the logarithmic probability, the support degree of all contexts to different candidate answers is obtained; the whole logarithmic probability vector is used as a support degree distribution condition of the context to the candidate answers; identifying a single piece of harmful information from the support degree distribution condition of the context to the candidate answers through a logistic regression model so as to filter wrong answers; according to the attack defense method provided by the invention, centralized injection of multiple malicious texts and sparse injection of a small number of malicious texts can be defended.

The invention discloses a fault self-adaptive diagnosis method for a rolling bearing for a rotating machine, and the method comprises the steps: firstly collecting a vibration signal in the operation process of the rolling bearing, and carrying out the preprocessing of the vibration signal, and obtaining the preprocessed vibration signal data; performing time sequence modeling on the preprocessed vibration signal data, extracting time sequence features, and generating a stage probability vector according to the time sequence features; mapping the stage probability vector into a stage embedded vector, introducing the stage embedded vector into a stage adaptive attention mechanism, and outputting an optimized feature vector; and finally, constructing a fault diagnosis network based on deep learning to perform fault diagnosis on the rolling bearing, outputting probability distribution of various faults of the rolling bearing, selecting the fault type with the maximum probability as a diagnosis result, and determining the severity of the fault type. According to the method, the fault stage can be automatically sensed, the evaluation standard can be dynamically adjusted, the method adapts to industrial data characteristics, the initial detection rate and the late false alarm rate are balanced, and therefore more accurate predictive maintenance is achieved.

A method of performing a probability tree analysis. The method includes identifying a plurality of nodes in a probability tree with each of the plurality of nodes having a probability vector. At least one node structural value is calculated for each of the plurality of nodes. The at least one node structural value quantifies an entropy of a subtree extending from a corresponding one of the plurality of nodes. The at least one node structural value is assigned to a corresponding one of the plurality of nodes. An analyzed probability tree is output including the at least one node structural value assigned to the corresponding one of the plurality of nodes.

The invention discloses a quantum error correction code decoding method, system and terminal based on momentum belief propagation, and the method comprises the steps: representing a prior message of the occurrence probability of each type of Follicle error of each quantum bit through a log-likelihood ratio, and obtaining an initialization message according to a prior probability vector, and updating messages transmitted to all adjacent variable nodes by the check node according to the initialization message to obtain an updated horizontal message, performing dynamic message updating on the posterior message of each variable node according to the updated horizontal message and the historical message to obtain an updated posterior message, and performing dynamic message updating on the posterior message of each variable node based on the updated horizontal message and the posterior message. And calculating an updated vertical message through a fixed suppression method, obtaining estimation of a real error according to a posterior message, and outputting a result. According to the method, a momentum item is introduced into message updating, a decoding algorithm based on momentum message propagation is obtained, and the decoding precision and the convergence rate of decoding on high-degenerate quantum codes are greatly improved.

A method for performing a probability tree analysis. The method involves identifying a plurality of nodes in a probability tree, where each node has a probability vector. At least one node structure value is calculated for each node. This at least one node structure value quantifies the entropy of a subtree extending from a corresponding node in the plurality of nodes. The at least one node structure value is assigned to a corresponding node in the plurality of nodes. An analyzed probability tree is output, containing the at least one node structure value assigned to the corresponding node in the plurality of nodes.

The invention discloses a power electronic transformer working mode recognition method and device, equipment, a medium and a product, and relates to the field of artificial intelligence, and the method comprises the steps: collecting original inductive current data during the operation of a power electronic transformer; performing adaptive segmentation normalization processing on the original inductive current data to generate a normalized inductive current sequence; calculating a wavelet packet energy entropy and a time domain differential entropy of the normalized inductive current sequence, and splicing the wavelet packet energy entropy and the time domain differential entropy into a two-dimensional fusion feature vector; inputting the normalized inductive current sequence and the two-dimensional fusion feature vector into a trained deep learning model to obtain a prediction probability vector; based on the prediction probability vector, the working mode category with the maximum probability value is selected as the recognition result, and the recognition precision of the working modes of the power electronic transformer can be guaranteed under the working conditions of high noise interference or rapid mode switching.

The invention relates to a model pruning method based on deep learning without participation of a real sample, which can solve the problem of lack of a current zero-sample category-based model pruning method. Comprising the following steps: firstly, preparing an image classification model, then creating an initial random image, normalizing the initial random image, inputting the normalized initial random image into the model, and obtaining an output mean value and a classification probability vector of each neuron; and jointly setting a loss function by using the classification probability vector, the target neuron output mean value and the output mean values of other neurons of the convolutional layer to which the neurons belong, then performing multi-round optimization on the depth dream, and finally performing de-normalization to obtain the final depth dream. And finally, inputting the deep dream into the model to obtain the prediction probability of the target category, and judging whether the neurons have important significance on the target category according to the prediction probability and a preset probability threshold so as to judge whether pruning is performed or not. According to the method, the importance degree of neurons can be judged without participation of a real sample, so that model pruning is performed.

The invention relates to an Internet of Things data man-machine interaction visualization system based on artificial intelligence, which belongs to the field of man-machine interaction technology, industrial Internet of Things and artificial intelligence technology and comprises a multi-modal data acquisition unit, a cognitive load state real-time inference module, a man-machine interaction interface self-adaptive generation module and a closed-loop regulation control module. The multi-modal data acquisition unit is used for acquiring multi-modal data related to the state of an operator in real time; the cognitive load state real-time inference module is used for inferring and generating a probability vector representing the current cognitive load state of the operator based on the acquired multi-modal data; according to the scheme, a complete feedback control loop is constructed by arranging the multi-modal data acquisition unit, the cognitive load state real-time inference module, the human-computer interaction interface self-adaptive generation module and the closed-loop regulation control module.

The invention relates to the field of temperature acquisition systems, and discloses a high-precision sampling processing method based on multi-model temperature sensors, and the method comprises the steps: obtaining an original ADC data block of a temperature sensor, and extracting a dynamic feature fingerprint; analyzing the dynamic feature fingerprints to determine a model probability vector characterizing a known working mode and a novelty score characterizing an unknown working mode; according to the model probability vector, the estimation results of the multiple extended Kalman filters are fused through an interactive multi-model algorithm, a temperature value is obtained, and a residual sequence is generated; and updating the system model online according to the novelty score or the residual sequence. The system comprises a data acquisition and feature extraction module, a sensor state sensing module, a self-adaptive temperature estimation module and a system online evolution module. According to the method, a closed-loop processing flow with state sensing and online evolution capabilities is constructed, so that the problem that the precision of a traditional fixed model is reduced under a complex working condition is solved.

The invention relates to the technical field of cable defect identification, in particular to a cable fault identification method based on a convolutional neural network. The method comprises the following specific steps: establishing a one-dimensional convolutional neural network model based on a transmission line model to predict a cable fault type and a fault point location; building a test platform to collect data of cables of different models and with a certain length, wherein the data is used for building a data set for model training; training the prediction model by using the data set, and iterating for multiple times until the target accuracy is met; reflected wave data of a to-be-detected cable are collected and input into the trained prediction model, and a fault type probability vector and the distance between the fault position and the initial end of the cable are output through the prediction model. The invention provides a one-dimensional convolutional neural network prediction model fusing multi-scale feature extraction and a time sequence attention mechanism, the model has good effects on cable fault identification and fault position prediction, and a cable fault identification model for different types and different defect positions is realized.

To efficiently generate a machine learning model for processing video data.SOLUTION: A training device 100 comprises: a to-be-trained model processing unit that uses a machine learning model to generate a prediction probability vector, a spatial token prediction probability vector, and a temporal token prediction probability vector of an object from training video data; a spatial teacher model processing unit that uses a spatial teacher model to generate a spatial teacher prediction class vector from image frames of the training video data; a temporal teacher model processing unit that uses a temporal teacher model to generate a temporal teacher prediction class vector from flow frames of the training video data; and a training unit that trains the machine learning model, on the basis of a first error between a movement label corresponding to the training video data and the prediction probability vector, a second error between the spatial token prediction probability vector and the spatial teacher prediction class vector, and a third error between the temporal token prediction probability vector and the temporal teacher prediction class vector.SELECTED DRAWING: Figure 1

The invention discloses an operation control method for a generative pre-training GPT model and an electronic device, and relates to the technical field of smart home / smart home, and the method comprises the steps: obtaining an input text through a GPT model; at least two probability vectors are generated according to the input text, the probability vectors correspond to a first word list, the first word list comprises an identification character and a plurality of first candidate characters, the identification character corresponds to a second word list, and the second word list comprises a plurality of second candidate characters; the statistical use frequency of the first candidate character is greater than the statistical use frequency of the second candidate character; according to the at least two probability vectors, at least two initial characters are obtained, each initial character corresponds to one probability vector, each initial character is derived from the first word list, and adjacent initial characters in the at least two initial characters have an incidence relation; and under the condition that the at least two initial characters meet the output condition, determining the at least two initial characters as the output characters of the GPT model.

The invention relates to the technical field of artificial intelligence, in particular to a method for training a knowledge base selection model and a user question processing method and device. Comprising the steps that multiple historical query questions of a user, reference replies of the multiple historical query questions and expected selection vectors of a knowledge base used for representing the multiple historical query questions and corresponding reference reply sources of the multiple historical query questions are obtained; determining a historical embedded vector of each historical query problem and a historical selection probability vector of each knowledge base under each historical query problem; determining a first loss value of each historical query problem according to the historical selection probability vector and the expected selection vector of each historical query problem; determining a total loss value of the knowledge base selection model according to all the first loss values; when the total loss value is smaller than the preset value, it is determined that training of the knowledge base selection model is completed, the model performance of the knowledge base selection model and the selection precision and selection efficiency of the knowledge base are improved, and more accurate replies are provided for subsequent user query.

Disclosed herein is a method and system for estimating the lane index for identifying the index of a current lane of a navigating vehicle. In one embodiment, the method includes determining a first score for lane model attributes and lane connectivity attributes using a transformer network by performing a first multi-head cross-attention mechanism on lane model and lane connectivity attributes. Further, the method includes determining a second score for fused sensor data attributes by performing a first multi-head self-attention mechanism on fused sensor data attributes. Subsequently, a third score is determined for the lane marking and the fused sensor data attributes by performing a second multi-head cross-attention mechanism on a lane marking and the fused sensor data attributes.The method further comprises determining an output probability vector by performing a second multi-head self-attention mechanism based on the first, second, and third evaluations, and identifying the index of the current lane of the vehicle based on the determined output probability vector.