844 results about "Prediction probability" patented technology

Provided are a method and an apparatus for constructing a road congestion prediction model, a device, a medium, and a product. A road traffic network is defined as a directed weighted graph. Historical dynamic traffic features of each road segment in the road traffic network are obtained as sample data, including recent dynamic traffic features and periodic dynamic traffic features. The sample data is input into a mixture of adaptive graph learners (MAGL) model for learning, and a probability prediction vector is output. The sample data is input into a trend expert model, and a trend distribution vector of a predicted probability of future traffic conditions is output. The periodic dynamic traffic features are fused to determine a periodicity prediction vector. An aggregated logit vector is obtained. An objective function is determined based on the aggregated logit vector. Congestion prediction training is performed to obtain a road congestion prediction model.

The invention discloses a systemic lupus erythematosus assessment method and device based on multi-modal medical map and medical knowledge fusion, terminal equipment and a medium, and relates to the technical field of medical artificial intelligence. The method comprises the following steps: acquiring multi-dimensional medical indexes of a patient, and constructing a multi-modal medical knowledge graph of the patient associated with the indexes; pre-training the self-supervised graph neural network to obtain a patient and medical index embedding vector; grouping and constructing a feature subset combination according to medical knowledge, and training and selecting an optimal sub-model; dynamically weighting and fusing the prediction probabilities of the sub-models to obtain a comprehensive prediction probability; and screening the key indexes based on the embedded vector and quantifying the activity evaluation contribution degree of the key indexes to obtain an activity comprehensive score. The systemic lupus erythematosus evaluation method improves the accuracy and robustness of preliminary evaluation of systemic lupus erythematosus, solves the problems of data missing and sample imbalance, and is explainable in output evaluation, adaptive to primary medical treatment and capable of supporting hierarchical diagnosis and treatment.

The invention discloses a flood disaster risk intelligent early warning method fusing multi-source information, and relates to the technical field of flood disaster prediction.The method comprises the steps that multi-source data of meteorology, hydrology, terrain, remote sensing, ground monitoring terminals and the like are obtained, and a unified data set is constructed through time synchronization, space resampling, abnormal value elimination and missing data filling; constructing multi-dimensional spatio-temporal features based on factors such as rainfall process, topographic features, surface coverage, drainage capacity, river network structure and soil humidity; inputting the spatial-temporal characteristics into a multi-source information fusion model, and predicting a flood occurrence probability in a future time period; flood risk indexes are generated according to the prediction probability, real-time monitoring and historical samples, and regional risks are divided into different grades; early warning information is automatically generated and issued according to the risk level; and the early warning result is fed back and corrected through field monitoring and patrol data, so that self-adaptive updating of early warning parameters and a model structure is realized.

The invention belongs to the field of bioinformatics, and relates to an improved integrated deep learning cell communication ligand-receptor interaction prediction method. The method comprises the following steps: firstly, carrying out extraction and dimensionality reduction on biological sequence features of a ligand and a receptor, and constructing multi-modal feature input; secondly, constructing an improved deep neural network branch, introducing a batch normalization layer and a Leaky ReLU activation function, solving the problems of gradient disappearance and neuronal necrosis, and improving regularization strength to prevent overfitting; meanwhile, an enhanced heterogeneous graph auto-encoder branch is constructed, the graph embedding dimension is remarkably expanded to improve the feature capacity, and full convergence of the model is ensured by increasing training rounds; thirdly, fusing the improved deep network with the prediction probability of a heterogeneous graph auto-encoder by adopting a weighted integration strategy; and finally, outputting a potential interaction relationship based on the fusion probability. By optimizing the architecture and the strategy, the prediction accuracy and robustness are remarkably improved, and a reliable tool is provided for analyzing a complex cell communication network.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a multi-mode early cognitive impairment screening method and system based on medical history information.The method comprises the steps that a first electrode plate and a second electrode plate are arranged at the forehead position and the occipital position of the brain of a user to be detected in advance; the third electrode plate and the fourth electrode plate are arranged on the left side of the brain side by side, and the fifth electrode plate and the sixth electrode plate are arranged on the right side of the brain side by side. And obtaining a disturbance coefficient set of the to-be-detected user under the reference detection mode combination, inputting the disturbance coefficient set into a pre-trained early cognitive impairment screening model, and predicting to obtain a prediction probability value of the early cognitive impairment of the to-be-detected user, and by adopting the scheme to screen the early cognitive impairment, the accuracy is high, the universality is strong, and the accuracy is high. The method can be applied to screening of people in communities, nursing homes and the like on a large scale.

The invention discloses a three-dimensional point cloud geometric information compression method based on implicit neural representation, and the method comprises the steps: constructing a trunk structure of an implicit neural network through a plurality of sine representation network layers which are connected in series, and introducing a variable-scale position coding mechanism on this basis, the method enables a network to obtain higher geometric reduction precision while keeping a compression ratio, and comprises the following steps: (1) inputting space coordinates of divided voxels into a position coding module with adjustable scale parameters; (2) feeding a coding result into an implicit neural network constructed by a network layer based on sine representation, and outputting the occupancy probability of the voxel through an activation function; (3) in a training stage, the model continuously optimizes parameters, so that the output probability distribution is highly consistent with a real occupied label; (4) after model training is completed, a method of combining an AdaRound second-order quantization optimization strategy and quantization perception training is introduced, network weight is finely adjusted, and quantization errors are reduced; (5) in a reasoning stage, judging whether the voxel is occupied or not according to a preset threshold value, and when the prediction probability exceeds the threshold value, regarding the voxel as occupied; and (6) all voxels judged to be occupied are aggregated, and reconstruction of the geometric structure of the point cloud is completed.

The invention relates to a generalized energy storage and micro-grid collaborative low-carbon operation method based on a dynamic electricity price strategy. The method comprises the following steps: S1, constructing a micro-grid operator unified scheduling model framework and initializing an electricity price signal; s2, constructing a demand response model driven by user satisfaction, and obtaining a user load curve; s3, generating an uncertainty random disturbance scene set based on a wind and light power prediction probability density function; s4, establishing a generalized energy storage model, and respectively constructing a virtual energy storage model of entity energy storage and EV aggregation in the multi-microgrid; step S5, constructing a low-carbon optimal scheduling model for the flexibility resource collaboration of the microgrid group; s6, solving the multi-microgrid scheduling optimization model in a wind-solar disturbance scene, and extracting a tail scene mean value as a risk constraint target of scheduling optimization based on CVaR; and S7, the micro-grid operator corrects the direction of the electricity price signal based on a CVaR result, and iteratively updates in combination with a differential evolution algorithm until the target value tends to be stable.

The invention relates to the field of identity association, and discloses a mobile terminal identity information association analysis method, system and device and a medium. The method comprises the following steps: acquiring a non-identification parameter set of a terminal to be judged and a stock parameter set of stock equipment; inputting the non-identification parameter set and the stock parameter set into a twin neural network model to obtain a prediction probability; when the prediction probability is greater than a preset probability threshold, determining that the to-be-determined terminal and the stock equipment are the same equipment, and associating the non-identification parameter set with the stock equipment; when the prediction probability is smaller than a preset probability threshold value, marking the to-be-judged terminal as a suspected new device, performing time sequence verification on the suspected new device and the stock device based on the dynamic time sequence parameter and the application behavior parameter, after verification is passed, re-judging that the to-be-judged terminal and the stock device are the same device, and if not, judging that the to-be-judged terminal and the stock device are the same device. And marking the to-be-judged terminal as a new device. According to the invention, the reliability and stability of association can be improved.

The invention discloses a hepatocellular carcinoma postoperative early recurrence prediction method based on multi-modal fusion. The method comprises the following steps: firstly, integrating clinical data of a training set, a preoperative enhanced CT image and a postoperative full-view digital pathological image, and carrying out standardized correction; then, traditional image omics features and deep learning features are extracted from the CT image, cell nucleus morphological features and tumor microenvironment spatial configuration features are extracted from the pathological image, and key feature signatures are screened out through a maximum correlation minimum redundancy algorithm (mRMR) and LASSO regression in combination with clinical features. And then carrying out progressive model construction by adopting an XGBoost algorithm, sequentially establishing a clinical single-mode model, an image single-mode model, a pathological single-mode model and a multi-mode fusion model, and explaining and visualizing the models by utilizing an SHAP value and a Grad-CAM technology. Finally, the performance of the model is evaluated in a multi-dimensional mode through internal cross validation, foresight and external independent validation, risk layering is carried out based on the prediction probability, and individualized postoperative management is guided.

The invention relates to a natural language processing technology, in particular to a fine empirical tracing and declaration verification method based on heterogeneous target fusion, which comprises the following steps of: retrieving candidate documents related to a to-be-verified declaration, and extracting candidate sentences related to the to-be-verified declaration; constructing a cascaded statement-sentence association probability distribution model and a statement-sentence set association probability distribution model; learning prediction probability distribution between the declaration and a single candidate sentence based on a declaration-sentence association probability distribution model; screening a plurality of candidate sentences with relatively high empirical values to obtain a candidate empirical set; on the basis of a statement-sentence set association probability distribution model, capturing a semantic synergistic effect between candidate sentences in the candidate empirical set, and calculating global category probability distribution of the statement to be verified and the candidate empirical set; jointly optimizing parameters of the two models by adopting a heterogeneous target; and on the basis of the two trained models, fine empirical cause tracing and declaration verification are realized. According to the invention, a visual and accurate viewpoint verification result can be provided.

The invention discloses an image classification method and system based on attribute-level optimal transmission alignment, and the method comprises the steps: dividing an input image into image blocks, adding a class mark vector, and forming the class mark vector and a visual feature vector through an encoder; secondly, based on the class mark vectors, visual attribute features are extracted, prompt statements are constructed, and text attribute features are obtained through a text encoder; then constructing an attribute-level cost matrix based on the visual attribute features and the text attribute features, and calculating to obtain attribute-level similarity between the visual attributes and the text attributes; according to the text attribute features, an attribute-level prediction probability is obtained through calculation, and a global feature prediction probability is obtained by applying a self-attention mechanism; and finally, carrying out weighted fusion on the attribute-level prediction probability and the global feature prediction probability to obtain a final category prediction probability. According to the method, collaborative learning of image semantics and attribute features is realized, so that image classification has higher accuracy and interpretability.

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 engineering performance intelligent detection, in particular to a concrete main tower circumferential prestressed reinforcement anti-cracking prediction method based on machine learning, which is characterized in that fiber grating strain sensors are arranged on the sections of main tower circumferential prestressed reinforcements, and strain data and environmental parameters in tensioning and operation stages are acquired; marking a cracking state to construct a sample data set; carrying out adaptive segmentation and abnormal value elimination operation on the strain sequence to realize data cleaning; constructing a cracking probability prediction model to process the cleaned data to obtain a cracking prediction probability; and calculating the total loss of the model, training the model, inputting newly collected data into the trained model, outputting a cracking prediction probability, and judging whether to carry out early warning or not. According to the method, online, real-time, intelligent and accurate prediction of the cracking risk of the circumferential prestressed steel bars of the concrete main tower can be realized, the monitoring false alarm rate is effectively reduced, and reliable technical support is provided for bridge operation maintenance and safety management and control.

The invention discloses a Mama dual-channel denoising network session recommendation method and system based on a hypergraph, and relates to the technical field of information retrieval, and the method comprises the steps: obtaining historical interaction data, and coding the historical interaction data to form article embedding; constructing a session hypergraph based on the session sequence, and constructing a sequence transfer hyperedge and a local co-occurrence hyperedge by taking an article as a hypergraph node; performing hypergraph attention aggregation on the session hypergraph to obtain high-order representation, and inputting a selective state space model to extract long-range dependence to obtain local channel session representation; constructing a session-level global graph, taking sessions as nodes, establishing connection according to a co-occurrence relation, and performing longitude sensitive pruning and entmax sparse attention aggregation to obtain global channel session representation; and fusing the two-channel session representation, and calculating a candidate item prediction probability to generate a recommendation result. By means of the technical scheme, collaborative modeling of the high-order relation in the session and association between the sessions is achieved, noise connection interference is restrained, and recommendation accuracy and stability are improved.

The invention discloses a cable tunnel autonomous obstacle avoidance inspection method, system and equipment based on multi-sensor fusion and a medium, and relates to the technical field of data processing, and the method comprises the following specific steps: correcting motion data collected by an inspection robot, carrying out pre-integration calculation according to the corrected motion data to obtain motion change data, and carrying out data processing on the motion change data; and predicting a frame state vector of the robot at the current moment, carrying out laser vision tight coupling optimization of the inspection robot, carrying out state prediction on a dynamic obstacle, carrying out collision risk assessment in combination with the frame state vector of the inspection robot, and establishing a path planning objective function to carry out real-time path re-planning. According to the method, the problem that the convergence speed and the real-time performance of a traditional algorithm are difficult to balance in a dynamic complex scene is solved through laser and vision tight coupling optimization. Through dynamic obstacle prediction, probability risk assessment and real-time path planning, an autonomous inspection system with active obstacle avoidance and real-time path generation capabilities is established.

The invention relates to the technical field of multi-modal data prediction, and discloses a fashion preference prediction method and device, and the method comprises the steps: obtaining multi-modal data comprising an image, a text and a user behavior time sequence, carrying out the feature extraction, and obtaining an image feature, a text feature and a time sequence feature; constructing spatial features based on the city embedded table and the corresponding regional culture label codes; aligning the dimensions of the spatial features and the time sequence features, and fusing the aligned spatial features and time sequence features by using a gating attention mechanism to obtain space-time fusion features; splicing the space-time fusion feature with the image feature and the text feature to obtain a multi-modal fusion feature; decoupling the multi-modal fusion feature to obtain a material decoupling feature, a style decoupling feature and a scene decoupling feature; and after feature splicing is carried out on the multiple decoupling features, the fashion preference prediction probability is obtained through Transform coding and MLP classification.

The invention discloses a motor imagery electroencephalogram signal decoding method, system and device based on a double-path hierarchical hybrid architecture. The method comprises the following steps: acquiring a multi-channel motor imagery electroencephalogram signal; extracting the preliminary spatio-temporal features through a convolution embedding module to obtain embedded features; the embedded features are input into a double-path hierarchical mixing module formed by stacking a plurality of feature processing sub-modules, feature extraction is performed on the embedded features layer by layer through a main path and an auxiliary path which are arranged in parallel, and different feature extraction strategies are adopted according to stacking hierarchies; in each layer, the dual-path output features are fused through an adaptive fusion mechanism to obtain depth features; and finally, outputting the prediction probability of the motor imagery category through a classifier module. The method effectively solves the problems that an existing method is high in calculation complexity, unbalanced in feature extraction and insufficient in robustness, and the decoding precision and efficiency of the motor imagery electroencephalogram signals are remarkably improved.

The invention provides a rheumatoid arthritis patient low muscle quality risk prediction method based on an uncertainty perception stacked meta-learning structure, and belongs to the technical field of machine learning. The method comprises the following steps: acquiring a rheumatoid arthritis data set; constructing a base learner set comprising a table Transform class network and a gradient boosting tree class model; splicing the out-of-fold prediction probabilities of all the base learners and the statistics thereof to form a meta-feature matrix; taking the meta feature matrix as input, and constructing and training a meta learner comprising a spectrum normalization multilayer perceptron and a random feature Gaussian process output layer; collecting to-be-detected data, inputting the to-be-detected data into the trained base learner set and the trained meta learner in sequence, and performing temperature scaling and beta-calibration on a prediction result; and taking the calibrated prediction probability as a final prediction result. Through multi-source data fusion, the defects of an existing model in the aspects of practicability, probability reliability and the like are overcome.

The invention discloses a micro-expression recognition method and device based on an attention mechanism and a medium. The method comprises the steps of preprocessing acquired micro-expression image data to generate a differential image and a vertex frame image; the differential image is input to an adaptive channel space interactive attention module to extract dynamic features of the micro-expression, the vertex frame image is input to an Euler feature weighting module to enhance static feature expression, and a deep residual network is used to extract static features of the micro-expression; the dynamic features and the static features are input into a difference statistics guide fusion module, and classification features are obtained by calculating feature difference statistics and generating a self-adaptive fusion weight; inputting the classification features into a classifier to obtain prediction probability distribution of micro-expression categories, and constructing loss function optimization model parameters based on the prediction probability distribution and real labels to obtain a trained micro-expression recognition model; and inputting the obtained to-be-recognized facial micro-expression sample into the micro-expression recognition model, and outputting a facial micro-expression recognition result.

The invention discloses a remote sensing image segmentation method based on multi-scale gating bottleneck convolution scanning, and relates to the field of remote sensing image segmentation, and the method comprises the steps: inputting a remote sensing image, and extracting initial features through an initial convolution layer; carrying out multi-level down-sampling processing on the initial features through an encoder, and carrying out feature enhancement on each level by adopting a multi-scale gating bottleneck convolution scanning unit in sequence to obtain multi-level encoding features; introducing a semantic bridging attention unit at the deepest layer feature of the encoder to carry out global semantic modeling and residual fusion to obtain an enhanced deep layer semantic feature; performing up-sampling processing step by step in a decoder, and fusing the up-sampled feature with the coding feature of the corresponding level to obtain a multi-level fusion feature; and inputting the fusion feature output by the decoder into the segmentation head to obtain a prediction probability graph. According to the method, the segmentation continuity and category discrimination robustness of slender targets, complex boundaries and small-scale ground features in the remote sensing image are improved.

The invention provides a large voice model training method and device, equipment and a medium, and the method comprises the steps: inputting a training voice data subset into a large voice model, and obtaining the prediction probability distribution of the training voice data subset outputted by a large language model module in the large voice model; according to the prediction probability distribution and the real probability distribution, an entropy weighted cross entropy loss function is determined, and the entropy weight of the entropy weighted cross entropy loss function is determined based on the distribution entropy of the prediction probability distribution at the current moment; and with the purpose of minimizing an entropy weighted cross entropy loss function, updating parameters of the large language model module and training a voice recognition tag of the voice data subset. In the training method of the large voice model, entropy weight is added in an entropy weighted cross entropy loss function, the problem of uncertainty of prediction probability distribution is solved to a certain extent, iteration is performed by using a self-feedback signal of distribution entropy based on prediction probability distribution, and the voice recognition effect of the large voice model is continuously improved.

The invention discloses a drug target prediction method based on fragment-level local and global feature fusion, and belongs to the technical field of computational biology and artificial intelligence drug design. Comprising the following steps: acquiring a medicine SMILES character string and a protein amino acid sequence; respectively segmenting the drug SMILES character string and the protein amino acid sequence to obtain a drug structure fragment sequence and a protein functional fragment sequence; and inputting the drug structure fragment sequence and the protein function fragment sequence into a pre-trained drug-target interaction prediction model to obtain a prediction probability of drug-target pair interaction. Compared with the prior art, the method has the advantages that convolution feature extraction, a multi-head attention mechanism and a gating fusion strategy are combined, an end-to-end DTI prediction framework is constructed, and the interaction between drugs and targets can be comprehensively mined.

The invention relates to the technical field of machine vision, in particular to an aluminum hose quality inspection method and device based on machine vision and computer equipment. The method comprises the following steps: configuring a defect-illumination response matrix, wherein the response matrix comprises detection confidence coefficient values corresponding to multiple preset defect types under multiple illumination conditions; acquiring a historical detection data stream on a current production line, and calculating a prediction probability vector of the multiple preset defect types at the current moment based on the historical detection data stream; inputting the prediction probability vector into an optimization model with a preset constraint condition, and combining the defect-illumination response matrix to solve a target illumination combination which maximizes a weighted detection confidence expectation value; and triggering a light source assembly to work according to the target illumination combination, collecting a surface image of the to-be-detected aluminum hose, and outputting a quality detection result based on the surface image. According to the method, the defects of the aluminum hose can be identified more accurately.

The invention relates to the technical field of river ice condition early warning, and discloses an ice plug prediction method based on fusion of a physical model and machine learning, and the method comprises the steps: obtaining and preprocessing the monitoring data of a to-be-predicted river reach; carrying out parallel calculation on a physical risk index based on a physical mechanism and a machine learning prediction probability based on data driving; performing weighted fusion on the index and the probability to obtain a basic risk index; then, performing multi-stage correction on the basic risk index based on the critical jamming condition and the river channel curvature to obtain a final risk index; and finally, determining and outputting an ice plug risk grade according to a preset threshold value. The mechanism definition of a physical model is combined with the ability of a machine learning model to capture a complex nonlinear relationship. By introducing critical jamming condition correction and river channel geometric feature reinforcement, the reliability of the prediction method under extreme working conditions is ensured, the applicability and prediction precision of the model to different river channel conditions are enhanced, and the comprehensive accuracy of a prediction result is improved.

The invention discloses a voiceprint recognition system attack resisting method and device, equipment and a medium, by constructing a shadow voice sample set and a substitution model, dependence on target speaker data and query capability is reduced, and privacy risk and query overhead are reduced; meanwhile, two types of substitution models are trained based on an information theory method, so that the mobility and stability of the models are improved, and the generalization ability is enhanced; besides, the attacker model adopts a specific architecture and combines an alternate training strategy, so that attack loss and prediction probability difference are optimized, imperceptibility and attack effectiveness are balanced, and tone quality reduction or attack failure is avoided; and finally, through systematic training and optimization, a more uniform evaluation caliber is expected to be provided, and more reliable guidance is provided for engineering landing and risk evaluation.

The invention discloses an intelligent management method and system for an electric power operation ticket based on deep learning, and the method comprises the following steps: collecting a historical compliance operation ticket text and an electric power industry specification text, and carrying out the text preprocessing operation; based on the preprocessed text, respectively constructing training samples for operation step prediction and operation sequence prediction; inputting the training sample into the constructed operation order prediction model, calculating step prediction loss, prediction sequence loss, sequence consistency score and prediction violation score, weighting the calculation step prediction loss, the prediction sequence loss, the sequence consistency score and the prediction violation score to serve as a total loss function, and training the model; inputting an operation task and an operation ticket text into the trained model, and calculating a semantic distance; and if not, obtaining the prediction probability and the sequence prediction probability of the operation task and the operation ticket text at the lexical element position, and positioning a wrong lexical element. According to the method, through joint operation step prediction, operation sequence prediction, sequence consistency evaluation and violation detection, high-precision prediction and error lexical element positioning of the electric power operation order are realized.

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 discloses a self-distillation map attention-enhanced multi-source remote sensing image fusion classification method. The method comprises the steps of obtaining hyperspectral data and laser radar data of a to-be-classified earth surface region; inputting the hyperspectral data and the laser radar data into a pre-trained semi-supervised self-distillation diagram attention enhancement network for feature extraction and fusion to obtain fused features; and classifying and outputting the fused features through a classifier to obtain fusion prediction probability distribution, and taking a category corresponding to a maximum probability value in the fusion prediction probability distribution as a final classification result. According to the method, complementary information of different modes can be effectively utilized, the distinguishing capability of the classifier is remarkably enhanced under the condition that dependence on prior knowledge is reduced, and ground feature types can be better recognized.

The invention discloses a two-stage space-time diagram network and multi-source information fusion time sequence probability prediction method and device, and relates to the technical field of artificial intelligence and big data analysis. The method comprises the following steps: acquiring historical wind speed sequence data and corresponding target variable sequence data; preprocessing the data, and constructing a training sample according to a preset time sliding window; constructing a time sequence probability prediction framework of the two-stage space-time diagram network and multi-source information fusion; based on the fluctuation correlation of the historical sequence data, constructing an adjacent matrix between nodes; inputting the training sample and the adjacent matrix between the nodes into a wind speed prediction model, training the model through a designed threshold perception loss function of a wind speed-power nonlinear relationship, and outputting a multi-node wind speed prediction probability of a first stage; and inputting the multi-node wind speed prediction probability and the multi-source environmental factor data into the gradient boosting tree model for second-stage prediction, and outputting a multi-node wind power prediction result. According to the invention, prediction errors can be reduced.

The invention discloses a substation safety behavior identification method and device based on a digital scene, and relates to the technical field of intelligent operation and maintenance. The method comprises the following steps: constructing a digital twinborn scene of a transformer substation; collecting multi-modal monitoring data, and performing safety behavior identification by using a safety behavior identification engine to obtain a key target and an illegal behavior prediction probability thereof; based on the digital twinning scene, according to system operation data collected in real time, an alarm strategy generation engine is used for strategy generation, and an optimal alarm strategy is obtained; according to the illegal behavior prediction probability and the optimal alarm strategy of the key target, triggering an alarm signal of a corresponding level, and performing visualization in a digital twinborn scene; and collecting a feedback signal for processing the alarm signal, and performing incremental optimization training on the safety behavior recognition engine and the alarm strategy generation engine according to the feedback signal. The problems of recognition lag, single data dimension and insufficient intelligence in the prior art are solved.