70 results about "Indicator matrix" patented technology

The invention provides a Beidou intelligent early warning analysis method and system applied to a drainage basin-level reservoir dam group, and the method comprises the steps: obtaining a dam body deformation time sequence record and a corresponding hydrological environment time sequence record which are collected by a monitoring system of the drainage basin-level reservoir dam group through a Beidou satellite positioning terminal; dam body deformation response mode self-encoding processing is carried out on the associated time sequence data set, a deformation response feature element set representing the dam body structure state is generated, the deformation response feature element set is input into a pre-constructed basin reservoir dam group topological relation graph, abnormal state propagation calculation is carried out through the state dependency relation between graph nodes, and the abnormal state propagation calculation is carried out. And obtaining a chain risk transmission path set, performing risk evolution trend deduction on dam nodes in each path based on the chain risk transmission path set, generating a risk evolution index matrix, and generating a threatened risk identification sequence for key dam nodes. According to the method, the structuralization and response operability of the early warning information can be improved, and the scientificity and practicability of early warning analysis of the basin-level reservoir dam group are comprehensively improved.

The invention relates to the technical field of multi-view clustering, in particular to a multi-view clustering method and device, and can solve the problem that the overall effect of an existing method in a large-scale clustering task is limited due to the fact that the existing method has problems in the aspects of calculation efficiency, robustness and multi-view information integration to a certain extent. The method comprises the following steps: dynamically learning an anchor matrix and a projection matrix for each view, and constructing a bipartite graph to generate a similarity matrix; calculating a graph Laplacian matrix based on the similarity matrix of each view, and extracting spectrum embedding; the spectrums of multiple views are embedded and stacked into a third-order tensor, and cross-view shared information is extracted by using a low-rank tensor constraint; multi-view atlas embedding is aligned through a spectrum rotation technology, and a discrete clustering indication matrix is directly output.

An example apparatus includes audience metrics collecting circuitry to access first audience metrics from a server, and access second audience metrics from the server, matrix building circuitry to build a matrix of the first audience metrics and the second audience metrics, missing values of the matrix corresponding to the second audience metrics, data transforming circuitry to transform the first audience metrics and the second audience metrics in the matrix, missing value calculating circuitry to determine imputed transformed values of the missing values using a recommender system, and the data transforming circuitry to recover imputed values of the missing values based on the imputed transformed values.

The invention discloses an asymmetric cross-modal hash learning method for potential structure information, and relates to the technical field of cross-modal retrieval. Analyzing a sample clustering structure, sample similarity information and semantic similarity information among the multi-modal samples to obtain a projection matrix of different modals, a clustering center of a projection space, a clustering indication matrix, a mapping matrix, a discrete hash code and a calculation formula of an auxiliary variable; updating the projection matrix of each modal, the clustering center of the projection space, the clustering indication matrix, the mapping matrix, the discrete hash code and the auxiliary variable by using an iteration strategy; determining a hash function of each modal according to the discrete hash code obtained by the last iteration and the multi-modal training sample; the hash function is used for performing similarity retrieval on the query sample in the multi-modal data. According to the method, the Hash code with higher discrimination can be obtained, so that the retrieval performance of cross-modal retrieval is improved.

The invention discloses a traffic flow prediction method and device based on a space-time diagram convolutional network, and the method comprises the steps: collecting road traffic flow data, and constructing a traffic congestion index matrix model; calculating a road traffic congestion index based on the collected road traffic flow data, and constructing a traffic flow adjacency matrix by using the road traffic congestion index; performing fast time convolution processing on the traffic flow adjacency matrix to generate a traffic flow data sequence containing spatio-temporal information; inputting the traffic flow data sequence into a space-time synchronization graph convolutional network model STSGCN, further learning a spatial topological structure, and deeply extracting space-time features by stacking multiple layers of space-time synchronization graph convolutional layers; and inputting the spatial-temporal characteristics into a wave loss function-based random vector function link network model Wave-RVFL optimized by an index and trigonometric function optimization algorithm, and predicting the traffic flow. The method can improve the precision and efficiency of traffic flow prediction.

The invention discloses a business flow site selection optimization method based on multi-source data fusion, and relates to the technical field of big data processing and multi-source data fusion, and the method comprises the following steps: obtaining multi-source data of a to-be-detected region, constructing multi-source features, carrying out the cross-modal fusion of the multi-source features through an adaptive weight distribution mechanism, and generating a comprehensive feature representation; based on comprehensive feature representation, passenger flow trends and causal contributions of the candidate points in different scenes are output through a causal inference method; taking the passenger flow trend and the causal contribution as input, establishing a multi-objective optimization model, performing solving, and outputting a comprehensive score and a sorting result of the candidate points; constructing an explanatory index matrix according to the comprehensive score and the sorting result in combination with a causal contribution result, and outputting a site selection scheme through a visual platform; according to the method, the problems of single data source and incomplete candidate point description in the prior art are solved by fusing the spatial geographic data, the space-time traffic data and the consumption behavior data.

The embodiment of the invention provides a public opinion evaluation method based on a multi-dimensional knowledge graph and related equipment, and belongs to the technical field of public opinion evaluation. The method comprises the steps that multi-source heterogeneous public opinion data are obtained and preprocessed, a multi-dimensional public opinion knowledge graph is obtained, and the multi-dimensional public opinion knowledge graph comprises entity nodes and association relations of users, topics, events, spaces and figures; based on the multi-dimensional public opinion knowledge graph, determining a hierarchical index system comprising a target layer, a middle layer and an index layer to obtain multi-dimensional public opinion index data; performing matrix standardization processing on the multi-dimensional public opinion index data to obtain a multi-dimensional public opinion index matrix; and inputting the multi-dimensional public opinion index matrix into a pre-trained convolutional neural network model for feature extraction, quantitative analysis and weight distribution to obtain a public opinion dynamic evaluation result. According to the embodiment of the invention, objectivity, accuracy and dynamic response capability of public opinion risk assessment can be improved.

The invention relates to an elastic balance area multi-objective planning method considering source-network load interaction, and the method comprises the steps: analyzing the real-time operation data of a power distribution network and the topological graph data of the power distribution network, and obtaining a source load feature matrix, an interaction mode matrix, and a final constraint set; constructing an optimization objective function set and a coordination mechanism set according to the source load feature matrix, the interaction mode matrix and the final constraint set; and establishing a comprehensive optimization model according to the optimization objective function set, the coordination mechanism set and the final constraint set to obtain a complete optimization model and a hierarchical structure matrix, and solving based on the complete optimization model and the hierarchical structure matrix to obtain a final optimization solution and a performance index matrix. According to the method, the uncertain influence caused by the random fluctuation power supply in the active power distribution network can be better processed.

The invention discloses a forest seedling raising period management optimization method based on an adaptive algorithm. The method comprises the following steps: S1, collecting multi-source environment data and seedling growth state data in a forest seedling raising process; s2, the original index data set is processed, a standardized index matrix is obtained, and a stage index evaluation input set is constructed; s3, constructing a multi-target periodic scheduling model based on the stage index evaluation input set; s4, introducing an entropy weight adaptive grey wolf optimization algorithm to solve the multi-target scheduling function set, and outputting a periodic scheduling optimal solution; s5, generating a seedling culture management execution strategy based on the period scheduling optimal solution; s6, forming a feedback monitoring data set by the environment feedback data collected in real time and the seedling feedback data, and generating an updated seedling management execution strategy; and S7, synchronizing the updated seedling culture management execution strategy to the seedling culture management platform. According to the invention, intelligent scheduling and adaptive management optimization of the whole forest seedling raising process are realized.

This invention belongs to the field of target detection technology, specifically relating to a method, system, and device for target area identification in complex electromagnetic environments. The method constructs a two-dimensional plane to form a planar coordinate system, using the center of the target area as the origin. The target area is discretized into a uniform grid, covering a square region centered at the origin. The inscribed circular region within the grid is considered the outer region, generating an outer region indicator matrix. This outer region indicator matrix is ​​then used to generate an arrival angle sector region indicator matrix, and further, an intersection region indicator matrix is ​​generated. Finally, the proportion of the arrival angle sector intersection region within the target area is calculated using the intersection region indicator matrix, and a judgment threshold is set. If the proportion is greater than or equal to the threshold, the target is determined to be within the target area; otherwise, the target is determined to be outside the target area. The system and device are used to implement this method. This invention significantly improves the accuracy, robustness, and computational efficiency of target area identification.

The invention provides a transformer operation and maintenance management method and system based on digital twinning, and relates to the technical field of power equipment operation and maintenance, and the method comprises the steps: carrying out the multi-source integration of transformer sampling data, and obtaining traceable state monitoring data through combining with a distributed block chain technology; adaptively decomposing the monitoring data according to frequency bands and extracting time-varying features, and introducing a cross-frequency domain feature fusion device based on mutual information maximization to analyze relevance of different frequency bands to obtain a state index matrix; constructing a state evolution diagram of the transformer parts, analyzing state propagation characteristics among the parts by adopting dynamic causal reasoning, extracting multi-scale time sequence characteristics, and performing layered interactive coupling to realize space-time conjoint analysis; and calculating the reliability index of each component based on probability distribution, tracing degradation time nodes and influence factors, and formulating a maintenance strategy. According to the invention, omnibearing monitoring and accurate prediction of the state of the transformer are realized, and the operation and maintenance efficiency and the equipment reliability are improved.

The invention discloses a forest stand attribute analysis method and system based on unmanned aerial vehicle remote sensing data feedback, and relates to the technical field of data processing, and the method comprises the steps: obtaining to-be-analyzed forest region data, querying the historical route data of an unmanned aerial vehicle based on position data, and determining remote sensing data collection equipment information and corresponding multi-source environment information; determining a forest stand attribute influence first sub-value of the to-be-analyzed forest; constructing an attribute analysis space coordinate system, determining to-be-analyzed remote sensing data corresponding to the coordinate points, and obtaining a to-be-analyzed forest stand attribute influence second sub-value; analyzing the remote sensing data to obtain a forest stand attribute index matrix; and based on the matrix and the to-be-analyzed forest stand attribute influence second sub-value, obtaining a forest stand attribute standard matrix and drawing up a forest management strategy. The system comprises an acquisition module and an unmanned aerial vehicle data and environment information association module. According to the method, systematic integration of multi-source data is realized, the forest stand attribute analysis precision is improved, and a scientific basis is provided for forest management.

An improved information entropy-based dynamic peak capacity analysis method and system for generator sets, characterized in that the method comprises the following steps: collecting performance indicator data of each generator set at each moment within a region, so as to form a performance indicator data matrix; normalizing the performance indicator data at a set moment and an offset value of the performance indicator data between adjacent moments, so as to respectively obtain a normalized performance indicator matrix and a normalized offset performance indicator matrix; calculating static and dynamic analysis peak capacities of each performance indicator of the generator set; and on the basis of the obtained peak capacities of the performance indicators and offset performance indicators, calculating an integrated peak capacity of each performance indicator of the generator set. By means of dynamically analyzing peak capacities at different moments, the present invention provides effective analysis rules for the power peak of generator sets, thereby determining the peak capacities of the generator sets at the current moment and different moments.

The invention discloses an economic development sharing degree evaluation method, device and equipment and a storage medium thereof, and relates to the technical field of power system optimization, and the method comprises the steps: collecting multi-source power data covering the power consumption behaviors and power supply service processes of urban and rural users, and carrying out the automatic arrangement processing of the multi-source power data, forming a standardized evaluation data set; generating an index matrix of evaluation indexes based on the standardized evaluation data set; based on the index matrix, determining the weight of each evaluation index by adopting a preset data driving mode; and based on the index matrix and the weight, determining an evaluation result of the economic development sharing degree of the power system. That is, a set of autonomous, dynamic, quantifiable and comparable evaluation mechanism is constructed in the power system, so that the economic development sharing effect of the economic development sharing degree in the power system is objectively depicted in multiple dimensions.

The invention discloses a new energy bearing capacity influence factor analysis method and system considering deep peak regulation, and belongs to the technical field of power system operation analysis and new energy consumption. The method comprises the following steps: for a power grid, generating deep peak regulation multi-gear evolution operation mode data, and carrying out static N-1 and transient N-1 security check on the evolution operation mode data; key operation indexes are extracted on the basis of the evolution operation mode data passing the security check, and an index matrix is constructed on the basis of the extracted key operation indexes; based on the index matrix, the influence weight of each index on the new energy bearing capacity is quantified, and the dominant factors influencing the new energy bearing capacity are identified. The method has important theoretical value and engineering guiding significance for improving the new energy consumption capability of the power grid and optimizing the deep peak regulation operation strategy.

The embodiment of the invention discloses a dynamic tracking detection method and device for disabled people. A specific embodiment of the method comprises the steps of obtaining basic information, a training guide information set and a working scene image sequence for a target disabled person; performing data standardization on the basic information and the training guide information set to obtain standardized basic information and a standardized training guide information set; constructing an evaluation index matrix; in response to determining that the disabled type meets a first preset type condition, performing target dynamic behavior detection on each working scene image in the working scene image sequence through a preset dynamic behavior detection model to generate a target behavior detection result sequence; and according to the evaluation index matrix and the target behavior detection result sequence, performing ability evaluation on the target disabled person to generate disabled ability evaluation information. According to the embodiment, the accuracy of a detection result can be improved.

The invention discloses a subjective and objective combination weighting evaluation method based on variance optimization, and the method comprises the steps: firstly evaluating the relative importance of each index, and obtaining the subjective weight of each index; constructing a sample-index matrix, calculating the information entropy value of each index, obtaining the weight of each index based on the information entropy, removing redundant indexes through the calculation of the index correlation degree and the correlation degree between the indexes, achieving the de-overlapping, calculating the absolute correlation degree of the indexes, and carrying out the normalization to obtain the objective weight of the indexes; setting subjective and objective weight coefficients, carrying out linear addition processing on subjective and objective weights, and on the premise that the sum of constraint condition coefficients is 1, enabling the variance to be as small as possible to obtain the optimal subjective and objective combined weight of each index; and judging subjective and objective weights of the indexes and the consistency of sorting arrays, and determining final weights of the indexes. On the basis of keeping the advantages of the original subjective and objective evaluation method, the advantage complementation of subjective weighting and objective weighting is realized, and the evaluation effect of the evaluation model is improved.

An embodiment of the present application provides a method for optimizing a non-orthogonal multiple access communication system based on a movable antenna, which is applied to a computer device. The method creates a channel model of the movable antenna communication system, creates a decoding strategy, calculates the minimum reachable rate from the base station to multiple user terminals through the decoding strategy, creates an optimization model for the minimum reachable rate, and splits the optimization problem of the minimum reachable rate into multiple sub-problems. The precoding matrix of the base station, the decoding indicator matrix, and the position vector of the movable antenna are optimized respectively. In an inner loop, the precoding matrix and the decoding indicator matrix of the base station are optimized through an alternating optimization algorithm. In an outer loop, the position vector of the movable antenna is optimized through an improved Hippo algorithm. The cyclic optimization of the sub-problems is realized through the inner and outer double loops. The optimized decoding indicator matrix, the precoding matrix of the base station, and the position vector of the movable antenna are obtained, the performance of the movable antenna communication system is optimized, and the minimum reachable rate between users is improved.

The application discloses a welding quality monitoring method of resistance spot welding data multi-neighbor feature analysis, which collects process data when the resistance spot welding is normal and implements standardization processing; obtains a regression coefficient matrix of the standardized data matrix; obtains a multi-neighbor relationship matrix through the regression coefficient matrix, further obtains a generalized eigenvalue problem and solves it; obtains a feature transformation matrix according to the solving result, further obtains a score matrix, a covariance matrix and a monitoring index matrix; obtains control upper limits of two monitoring indexes according to the monitoring index matrix; implements standardization processing on the latest process data; obtains a score vector and an error vector of the standardized data vector according to the feature transformation matrix, further obtains the two monitoring indexes; determines whether the welding quality is normal or not by judging whether the two monitoring indexes are within the control upper limits; and the method has the advantage that the reliability of the welding quality monitoring can be effectively improved.

The application discloses a traffic accident severity analysis method based on a hierarchical clustering evaluation model, and comprises the following steps: S1: selecting the number of light injuries, severe injuries and deaths in the accident as evaluation indexes, and then constructing an evaluation index matrix according to the following equation. The application aims to classify the severity of traffic accidents in a region every day, and analyze the time variation and influence factors of the severity, so that the corresponding comprehensive severity variation and influence factor analysis conclusion can be obtained, and the severity variation and influence factor analysis conclusion can be used as reference data for traffic management strategies under major public time.

The application relates to the technical field of vehicle network cloud service, in particular to a quality evaluation method and system for vehicle network cloud service; the method mainly comprises the following steps: constructing an index matrix based on unified measurement, defining index names, constructing an index model, extracting cloud service data, and analyzing and calculating output results according to the constructed index model; in the scheme, the measurement is unified, demanders can use the same measurement index to evaluate the quality of different cloud services, and the quality of different cloud services is evaluated from multiple dimensions, systems and business dimensions, so that the evaluation accuracy is improved; in the scheme, a universally applicable index matrix is constructed, the cloud service is measured according to a unified standard from multiple angles, cloud service log data is processed by using a data warehouse technology, the accuracy of index calculation is improved, and the index calculation process is simplified.

The invention relates to the field of data processing of disaster risk management and geographic information, in particular to a flood risk zoning method based on multi-source data entropy correction G1 empowerment. The method comprises the steps of obtaining multi-source data and performing metadata management to generate an original index set; performing quality control and aperture alignment to form a quality control index matrix; calculating entropy values based on the quality control data, and constructing an entropy difference set with difference measurement; generating an adjacent ratio mark set through stable item addition, approach threshold rollback, upper and lower limit truncation and monotone constraint; generating a weight vector by adopting hierarchical grouping and G1 inverted recursion; carrying out weighted summation to obtain a comprehensive risk scoring sequence; generating a five-level risk tag through K-means clustering; and finally, outputting a result data packet and a risk zoning map through GIS association visualization. According to the method, objectification, automation and traceability of weight distribution are realized through an entropy correction G1 weighting method, and the accuracy and practicability of flood risk zoning are remarkably improved.

The invention discloses a multi-dimensional fusion power system toughness index and evaluation method and device. The method comprises the following steps: collecting power grid data of a to-be-evaluated power grid and determining extreme event characteristics; obtaining a power grid function change curve through a power grid simulation or fault deduction method based on the power grid data and the extreme event characteristics according to pre-constructed power grid function measurement containing a power grid structure; calculating a power grid toughness index matrix according to the power grid function measurement and the power grid function change curve; and weighting the power grid toughness index matrix to obtain the power grid staged comprehensive toughness of the to-be-evaluated power grid.

The invention discloses a real-time lightweight backdoor detection method and system based on deep neural network parameter analysis, and belongs to the field of deep learning safety. In the DNN model, a feature extractor and a classifier are two key parts for processing a clean sample and a backdoor sample, and the last layer of the model has the most direct and key influence on the performance of backdoor attack. According to the method, 62 parameter anomaly indexes are defined, and contributions of the indexes to backdoor detection of different data sets and model architectures are analyzed through a supervision or non-supervision method. Finally, the method is configured for each model-data set pair to identify the optimal index combination, the index matrix of the model is extracted as an anomaly detection score, and the back door model and the clean model are distinguished by measuring the cosine similarity between the score distribution of the model and the score distribution of the clean model. And the category with the maximum score index is identified as the poisoning category.

The invention relates to a multi-mode self-adaptive oral English training method, system and device and a medium. The method comprises the following steps: acquiring multi-modal data; extracting features of spoken language ability corresponding to the multi-modal data to obtain a feature vector group; performing standardization processing on the feature vector group to obtain an index matrix; inputting the index matrix into a pre-trained attention mechanism model to obtain a spoken language ability portrait corresponding to the user; taking the spoken language ability portrait as an index, and matching corresponding teaching resources in a preset training resource library to obtain modular training resources; the modular training resources are courses corresponding to capability defects of the user; and generating a training scheme of the corresponding user based on the modular training resources. By adopting the method, multi-modal data can be introduced to carry out quantitative evaluation on the spoken language ability of the user, and the reasons of errors in the spoken language dialogue process of the student can be effectively judged.

This application provides a method and apparatus for interlayer identification, relating to the field of oil reservoir resource development technology. The method includes: constructing an index matrix based on first index data from M sampling points in a well logging curve; obtaining the weight corresponding to each type of first index data; weighting the index matrix according to the weights to obtain a result matrix composed of second index data; determining the interlayer development probability of each sampling point based on the maximum and minimum values ​​of each type of second index data in the result matrix and N types of second index data corresponding to each sampling point; obtaining the development probability of each of K well logging curve units in the well logging curve based on the interlayer development probability of each sampling point; and determining the existence of an interlayer if the development probability of any well logging curve unit is greater than a preset threshold. Identifying interlayers quantitatively based on the well logging curve can improve the efficiency of interlayer identification.

The application relates to the technical field of intelligent garbage sorting, and discloses a garbage sorting method based on artificial intelligence. The method performs heterogeneous data fusion on a multi-source garbage feature data set through a semantic alignment module, unifies feature dimensions and eliminates redundant items, and outputs a standardized feature set. Spatial position identifiers and physical state identifiers of garbage entities are extracted, a migration learning model is used to dynamically divide time slices to generate a periodic garbage flow data set. The starting point and ending point coordinates of sorting are converted to a three-dimensional space grid to construct a migration trajectory topology graph, the flow data and the physical state identifiers are combined, the sorting path is analyzed according to material categories, the material flow intensity in a specific period in a target area is quantified, and a dynamic index matrix is generated. The method realizes spatiotemporal correlation analysis of multi-modal garbage data, and improves the adaptability of sorting path planning in a complex environment.

This invention discloses a nonlinear control law evolution method based on a large language model and control effect evaluation index, including initializing the control law; converting the weights of the control law operators into sampling probabilities; sampling each control law operator according to the probability distribution of the control law operators in the large language model to generate a control law formula; inputting the set of control law formulas into an aircraft dynamics simulation model to obtain a tracking error sequence and determining the evaluation index matrix of the tracking error sequence; analyzing the relationship between the control law operators and the evaluation index and outputting the update gradient of each control law operator; updating the weights of each control law operator and generating multiple new control law operators, and initializing the weights of the new control law operators; ending the iteration when the number of iterations reaches the maximum limit and outputting the control structure and control parameters of the control law, otherwise resampling and iterating; this invention solves the problem that large language models cannot be embedded in the dynamics simulation environment and evaluate the effect of control laws.

The invention discloses a data evaluation model training method, a data evaluation method, a data evaluation device and electronic equipment, and relates to the technical field of computers, the data evaluation model training method comprises the following steps: preprocessing a plurality of historical index matrixes corresponding to a plurality of cities, and predicting to obtain index values in a first time period; reconstructing each historical index matrix based on the predicted index values in the first time period to obtain at least one first index matrix; performing clustering processing on the index data in the at least one first index matrix to obtain first clustering graphs corresponding to the at least one first index matrix; calculating an index weight in each first clustering graph through an entropy evaluation method to obtain a comprehensive score of the plurality of cities; and inputting the at least one historical index matrix corresponding to the plurality of cities and the comprehensive scores of the plurality of cities into a preset model for training to obtain an optimized evaluation model.

The invention belongs to the technical field of power distribution line monitoring and analysis, and particularly discloses a power distribution line loss abnormity analysis and search method, device and equipment and a medium, and the method comprises the steps: obtaining the power grid operation data of a target power distribution line; determining a line section with suspected line loss abnormity according to the power grid operation data of the target distribution line; in the line section with the suspected line loss abnormity, constructing a characteristic index matrix of each node in the middle through multi-source data fusion; based on the feature index matrix of each intermediate node, calculating the weight of each feature index by adopting an improved TC-EWM; determining the detection priority of each intermediate node based on the weight of the feature index; and detecting based on the detection priority of each intermediate node, and determining a node with abnormal line loss according to a detection result. According to the invention, rapid positioning and efficient treatment of line loss abnormity can be realized, the precision and economical efficiency of line loss treatment are improved, and the operation safety and intelligent level of a power distribution system are improved.