143 results about "Spectral clustering algorithm" patented technology

The invention discloses a computer task scheduling method based on artificial intelligence, and the method comprises the following steps: 1, data collection: employing a double-flow feature fusion mechanism, and generating global feature representation containing long-term dependence and an instantaneous state; step 2, generating a global optimization scheduling strategy: constructing a hierarchical federal reinforcement learning system, dividing a cluster into a plurality of super nodes through an enhanced spectral clustering algorithm, independently training a Dueling DQN network by each super node, performing global strategy cooperation by adopting Shapley value weighted aggregation and differential privacy protection, and generating a scheduling strategy of global optimization; distilling a global strategy into a lightweight decision tree through a strategy distillation technology, and deploying the lightweight decision tree to a physical node; 3, task priority control and elastic resource allocation are carried out, wherein elastic control over resource allocation is carried out through a dynamic time slice bank mechanism; and 4, self-adaptive evolution: establishing a closed-loop optimization system, and carrying out strategy self-evolution by adopting a double-layer optimization architecture.

The invention discloses an automatic flow chart layout method fusing semantic analysis, which relates to the technical field of automatic flow chart layout, and comprises the following steps of: inputting a node set with a business ring dependence condition into a time sequence conflict analysis engine, and combining a semantic vector and a time sequence vector to obtain a time sequence conflict analysis result; calculating a time sequence conflict index of the annular dependency set by adopting a weighted path consistency check algorithm so as to determine a time sequence conflict degree under the condition that the nodes have service annular dependency, and generating corresponding conflict description data; and inputting the conflict description data and the semantic vector into a conflict perception clustering optimizer, introducing a time sequence conflict penalty term into a clustering objective function, and performing cluster boundary adjustment on the annular dependency set through a spectral clustering algorithm so as to adjust a semantic clustering structure according to a determination result. According to the method, the problem that a semantic clustering structure cannot be optimized in combination with a time sequence conflict under business annular dependence is solved, and the effects of conflict accurate identification, clustering boundary dynamic adjustment and layout saliency enhancement are achieved.

The invention discloses an intelligent early warning analysis platform based on management and control of a product life cycle, and the platform comprises a data collection and preprocessing module which is used for collecting multi-source heterogeneous data and carrying out the preprocessing of the data, and obtaining a standardized input data set; the spectral clustering grouping module is used for carrying out unsupervised grouping by adopting a spectral clustering algorithm and generating a grouping label and a mapping result; the abnormal sample counting and generating module is used for counting the number of abnormal samples and data distribution of each data sub-group and generating abnormal sample data matched with sub-group characteristics through a stable diffusion model; the training and risk detection module is used for constructing a training data set and training a risk identification and anomaly detection method; and the intelligent early warning and response module is used for automatically triggering early warning for the detected abnormal data and linking platform management and control. According to the method, intelligent grouping, abnormal sample enhancement and efficient risk early warning analysis of product full-life-cycle multi-source heterogeneous data are realized.

The invention discloses an internet of things-based amusement equipment light dynamic control method and system, and the method comprises the steps: collecting the acceleration, angular velocity and coordinate data of equipment, carrying out the wavelet denoising and Z-score standardization processing, and constructing a dynamic topology model based on a graph attention network and a graph convolution network. A spectral clustering algorithm is utilized to carry out space partitioning on equipment nodes to generate a topological sub-graph, and an LSTM network is combined to predict an equipment motion track and generate a continuous track sequence. Track correlation features are extracted through a graph attention network, a genetic algorithm is fused to optimize and generate a light conversion sequence, dynamic time warping alignment timestamps are synchronously adopted, and phase synchronization of the light sequence and the motion track is ensured. And finally, brightness, color and flicker frequency are adjusted in real time based on a PID control algorithm, and a closed-loop feedback mechanism is formed. According to the method, high-precision dynamic matching of light and equipment movement is realized, and immersive experience and visual interactivity are improved.

The invention discloses an electric field analysis and optimization method for a transformer withstand voltage and partial discharge test platform, and the method comprises the steps: building a digital model of the test platform, obtaining a preparation technology standard and historical manufacturing measurement data of a target test platform, defining an influence variable which influences the distribution of an electric field, simulating the distribution of the electric field under different preparation differences, and carrying out the optimization of the electric field. Establishing an electric field simulation database; identifying test platform electric field sensitive areas under different preparation differences by adopting a spectral clustering algorithm based on the electric field simulation database, constructing a knowledge graph, constructing a sensitive area identification model based on a graph neural network, and performing model training through the knowledge graph; structural features and preparation features of the to-be-analyzed test platform are obtained, a local sensitive area of the to-be-analyzed test platform is recognized in the sensitive area recognition model, and if the local sensitive area exists, an optimization scheme is formulated for test platform optimization assistance, so that the design reliability and operation safety of the transformer test platform are improved.

The invention discloses an intelligent set top box video processing system based on multimode decoding, which relates to the field of terminal video processing, and comprises the following steps: an original data stream processing module outputs a scene complexity grading label through a spectral clustering algorithm; the video metadata analysis module is used for analyzing metadata of an original video stream in real time and identifying a content type identifier; the edge-cloud cooperative decoding module allocates the high-complexity video frames to edge nodes for decoding, allocates the low-complexity video frames to a local terminal for decoding, and outputs a decoded video frame sequence; the video restoration module detects and restores picture distortion caused by transmission frame loss or signal interference; the dynamic code rate control module dynamically distributes the code rate weight of each block-level area according to the scene complexity grading labels; and the energy efficiency self-optimization module dynamically controls the states of the heating sheet and the idle decoding channel to adjust the core voltage and the decoding frame rate of the CPU. The method has the advantages of low power consumption, strong adaptability to weak network smoothness through multi-scene high-definition cloud edge collaboration.

A honeycomb-shaped active power distribution network topological structure selection and optimization method based on a graph theory comprises the following steps: taking line impedance and rated apparent power as constraints, constructing a multi-target weight factor, adopting an improved spectral clustering algorithm to cluster and distribute power grid nodes, and transiting a micro-grid group after cluster division into a honeycomb-shaped active power distribution network; abstracting each micro-grid as a honeycomb distribution network single node, constructing a connection relation mathematical representation of the HSPH and the micro-grid, and realizing sparse matrix method modeling of the honeycomb active power distribution network; optimizing a site selection strategy of the HSPH by taking a maximized HSPH return on investment index and a system stability index as a target function; and optimizing the energy storage capacity of the HSPH based on the net load time sequence data of the micro-grid group, and finally generating an optimal topological structure of the honeycomb-shaped active power distribution network. And through honeycomb topology and HSPH addressing and sizing optimization, the new energy consumption capability of the power distribution network is improved, and the economy, reliability and flexibility of the power distribution network are improved.

The invention discloses an analog IC synchronization test optimization method and system based on crosstalk simulation. The method comprises the following steps: firstly, acquiring position layout data and test signal data of each analog IC in a synchronous test process of the analog ICs, determining electromagnetic crosstalk information by combining test signals, and constructing a crosstalk schematic diagram according to the electromagnetic crosstalk information and the position layout data; clustering the schematic diagrams by using a spectral clustering algorithm, and identifying a crosstalk path in the synchronization test; designing an electromagnetic crosstalk signal isolation barrier based on the schematic diagram and the crosstalk path, deploying the electromagnetic crosstalk signal isolation barrier in a test environment, performing simulation evaluation on an independently tested analog IC by constructing and injecting a crosstalk simulation signal, and judging an isolation effect; and optimizing the isolation barrier according to an evaluation result to obtain an optimization scheme. According to the method, the electromagnetic crosstalk path in the synchronous test process can be effectively identified and isolated, and the accuracy and reliability of the simulation IC test are improved.

The invention discloses a topological communication coverage path planning method and system of an unmanned ship cluster for wide-area target search and detection, and belongs to the technical field of multi-agent collaborative path planning. According to the method, turning frequency optimization is converted into a minimum rectangular coverage problem, an integer programming model is established, a layered relaxation integer programming acceleration strategy based on maximum extension candidates is proposed, and the bottleneck of model expansion and calculation efficiency caused by candidate solution space explosion in a large-scale scene is broken through; then, constructing a rectangular topological connected graph, designing an improved spectral clustering algorithm fusing load balancing constraint and subgraph connectivity guarantee, realizing efficient and balanced distribution of regions in a complex environment, and ensuring internal connectivity of subregions; and finally, providing a dynamic pruning optimization method for the minimum corner spanning tree based on strong and weak connection characteristics, iteratively evaluating the corner cost and removing redundant edges, and finally generating a coverage path spanning tree with the minimum corner cost, thereby providing an accurate and efficient full-coverage operation path for scenes such as regional exploration and environment monitoring.

A power distribution network voltage cooperative control method considering a cluster division-optical storage all-in-one machine comprises the following steps: step 1, combining an electrical modularity index, a balance index and a membership index to construct a cluster comprehensive division index; 2, taking the comprehensive division index as a convergence condition for dividing a cluster region, and solving an optimal cluster division result of the power distribution network by utilizing an improved spectral clustering algorithm; 3, with the minimum power distribution network node voltage deviation punishment cost, the minimum network loss cost and the minimum power generation loss cost of the optical storage all-in-one machine as the target, a power distribution network voltage optimization model containing the optical storage all-in-one machine is constructed; and 4, based on the optimal cluster division result, solving the power distribution network voltage optimization model of the optical storage all-in-one machine through a cluster voltage optimization control strategy, optimizing active power, reactive power and voltage of the power distribution network, and improving power flow distribution of the whole network. The voltage regulation task sharing and power balance are realized, the voltage regulation effect is more reasonable, and the voltage out-of-limit problem under the random fluctuation of the source load power can be effectively solved.

The invention provides a sewing thread defect classification method based on multi-scale fusion and adaptive learning, which solves the technical problems of extremely unbalanced data and low expert labeling efficiency in industrial quality inspection, and can be extensively applied to surface defect detection tasks in multiple manufacturing fields such as spinning, packaging and electronics. According to the method, four core modules of multi-scale depth feature extraction, dynamic loss prediction, adaptive spectral clustering and intelligent pre-classification are creatively fused, and multi-level semantic features are extracted through a multi-scale feature pyramid network; designing a dynamic weight loss prediction network, and accurately evaluating the sample information amount; constructing an adaptive spectral clustering algorithm, and adaptively adjusting clustering parameters; a multi-feature fusion pre-classification algorithm is combined to assist experts in efficient labeling. According to the method, the defect classification precision can be remarkably improved in a few-sample scene.

The invention relates to the technical field of Web security and artificial intelligence, and discloses a Web reverse analysis method based on a large model and dynamic feedback, which comprises the following steps: constructing an AST-LLM joint analysis engine, defining an AST node importance evaluation formula, carrying out weight quantification on AST nodes, and screening out key functions or code snippets in reverse analysis; semantic analysis is carried out through LLM in combination with context, and candidate names of confusion variables are generated; semantic similarity calculation is carried out on the key nodes screened out through weight quantification and candidate variable names generated by LLM, and semantic grouping is carried out on confusion variables associated with AST nodes through a spectral clustering algorithm; a code structure is optimized through AST node remapping, and codes are corrected through real-time interaction; a closed-loop optimization system of LLM and log instrumentation is constructed, and an optimization engine is dynamically fed back; the method has the advantages that multi-state confusion is resisted, the semantic consistency of reverse output and original logic is ensured, the problem of reverse analysis distortion in a dynamic confusion scene is solved, and the readability, the performability and the maintainability of codes are kept.

The invention discloses a data analysis management method and system based on ocean network security, and relates to the technical field of ocean network security, and the technical key points comprise the following steps: collecting the flow and log data of multi-source ocean network equipment in real time, constructing a dynamic heterogeneous data stream, and extracting a space-time correlation feature tensor; behavior topology modeling is carried out on the space-time correlation feature tensor, a network behavior dynamic topological graph is generated, and topology stability measure is calculated; abnormal behavior detection is carried out based on topological stability measurement, an abnormal behavior cluster is identified through a multi-scale spectral clustering algorithm, and a high-risk threat area coordinate set is generated; the technical effects are that the dynamic topology modeling adapts to the characteristics of ocean network equipment movement, communication intermittence and the like, and the environment adaptability is improved; and manifold mapping and adaptive clustering are combined, so that the anomaly recognition accuracy is improved, the missing report rate is reduced, and accurate threat positioning is realized.

The invention relates to the technical field of indoor layout design, and discloses an indoor layout design system based on user behavior data analysis. The system comprises a user behavior data acquisition module which captures and transmits a user movement track coordinate sequence, residence time distribution data and equipment interaction event records in real time through a multi-sensor network; the layout analysis module constructs a dynamic layout model by using a decision tree integration algorithm, and outputs theoretical layout parameters including a space occupancy rate and functional region division; the layout difference detection module performs multi-dimensional comparison of space overlapping degree, path conflict index and regional utilization efficiency difference on theoretical and actual layout parameters to generate a difference index vector; the problem region positioning module identifies a layout abnormal region through a spectral clustering algorithm based on the difference index vector and predefined spatial topology information; and the adaptive design strategy generation module automatically generates a layout adjustment strategy instruction according to the abnormal region position and the attribute data.

The invention discloses a semi-supervised image clustering method based on adaptive graph learning, and the method comprises the steps: iteratively updating a sparse representation matrix and a paired constraint matrix until the value of a target function is unchanged or reaches a maximum iteration number; taking the updated pairwise constraint matrix as an input similar matrix, calling a spectral clustering algorithm to divide image sample data into a plurality of sample groups to finish clustering output; wherein the objective function is a weighted sum of a propagation consistency error based on a sparse representation matrix and a paired constraint matrix, an image sample data reconstruction error based on the sparse representation matrix, an L1 norm of the sparse representation matrix and a matrix correlation error based on the sparse representation matrix and the paired constraint matrix. The invention further discloses a system, a computer storage medium and a program for implementing the method. According to the method, the problem of separation of similar graph learning and constraint propagation can be solved, and the image clustering performance is improved.

The invention provides a seismic fault identification method and system based on multi-attribute collaboration and spectral domain graph enhancement, and relates to the technical field of seismic exploration, and the method comprises the steps: carrying out the all-directional dip angle and azimuth angle scanning of a seismic data volume, extracting multi-scale guide field information through combining structure tensor decomposition, and executing anisotropic diffusion filtering; calculating characteristic value distribution through a characteristic value coherence algorithm based on the filtering data volume, and determining the structural consistency difference of adjacent seismic traces to obtain a fracture coherence attribute data volume; carrying out azimuth gather sorting and pre-stack time migration processing on the seismic data volume, extracting seismic wave dynamic response characteristics, carrying out Fourier series expansion on the azimuth change rate to obtain a crack indication information data volume, splicing the data volume and executing multi-scale three-dimensional convolution solution, and constructing a spatial dependency graph through a spectral clustering algorithm; spectral domain enhancement features are obtained through spectral domain graph transformation and frequency selective filtering reconstruction, and morphological connectivity analysis is executed to obtain a crack prediction result.

The invention discloses a degradation dominant characteristic factor identification method for a river ecosystem, and particularly relates to the technical field of ecological environment monitoring and system modeling. A space-time tensor model of multi-source ecological observation data is constructed, multi-dimensional dynamic principal component analysis and a spectral clustering algorithm are fused, a high-contribution-degree feature sequence is extracted, and a multi-scale ecological feature association map is constructed; on the basis, interaction weights among ecological variables are defined, an ecological degradation path network is constructed, and an intermediary variable and a response variable are identified. The explanatory force of each feature is further evaluated by adopting a Shapley value method, a dominant feature importance degree matrix is formed, and a driving response model is established in combination with historical intervention measure data, so that scene optimization of a repair strategy is realized.

The invention discloses a health perception adaptive control method and system for cooperation of multiple energy storage converters, and relates to the technical field of adaptive adjustment. The method comprises the steps of performing feature extraction based on operation state data and operation environment data of N energy storage converters at M time points to obtain a corresponding comprehensive feature set; inputting the comprehensive feature set into a state diagnosis model to obtain an energy storage converter state label; the method comprises the following steps: constructing an undirected graph by taking an energy storage converter as a node and an electrical connection line as an edge, and dividing the undirected graph into G sub-graphs based on topological connection closeness and state similarity by adopting a spectral clustering algorithm; extracting a topological characteristic parameter set from each sub-graph; traversing each sub-graph, judging whether adjustment is ended or not according to a state balance degree threshold value, if adjustment is not ended, identifying an unbalanced scene according to topological characteristic parameters, and executing adaptive control for different scenes; according to the invention, health perception self-adaptive power distribution of cooperation of multiple energy storage converters is realized, overload and idle coexistence are avoided, and circulation oscillation is suppressed.

This invention discloses a multi-view subspace clustering method based on diversity graph fusion. The method comprises the following steps: Step 1: Acquire multi-view data and perform preprocessing; Step 2: By adding regularization terms for multi-view consistency and diversity, introduce self-expression learning to explore the intrinsic structure of the multi-view data, and introduce low-rank and sparse constraints on the consistency expression matrix, thereby obtaining a highly reliable and robust similarity matrix; Step 3: Using an induced self-weighting approach, fuse the view similarity matrices obtained in Step 2 to form a final consistent similarity matrix, which serves as the input of a spectral clustering algorithm and outputs the clustering results. This method can improve clustering performance and achieve optimal clustering results.

The invention provides an urban building group three-dimensional function partitioning and floor population dynamic distribution method and device, and relates to the technical field of urban space information processing. The method comprises the steps of constructing semantic feature vectors of multi-source POI data based on a BERT model and domain specific vocabularies to generate fusion data; building a floor position probability distribution model according to a three-layer progressive reasoning structure and a reasoning technology; carrying out three-dimensional function partitioning according to the function coupling strength and an improved spectral clustering algorithm; building-level demographic data are generated, and a differentiation attraction model is constructed; and solving the floor population distribution constraint optimization model according to the improved Haff model to obtain population dynamic distribution of each floor. According to the method, granularity limitation of traditional function division is broken through, floor-level refined three-dimensional function division and population dynamic distribution prediction are realized, and important decision support is provided for applications such as commercial site selection, emergency evacuation path planning, public service facility configuration optimization and the like in smart city construction.

This invention relates to the field of power grid security technology and discloses a method for searching critical transmission sections based on normative tangents and security risks, comprising: Step S1: Importing historical power grid operation data from the power grid operation database; Step S2: Generating operation scenarios using the Latin hypercube sampling method; Step S3: Calculating power flow and node voltages using the Newton-Lawrence method; Step S4: Constructing a branch weight model; Step S5: Constructing a power system graph theory model based on spectral graph theory, the power system graph theory model including the number of network nodes, the number of branches, and weights; Step S6: Partitioning the power grid topology map using a spectral clustering algorithm based on normative tangents; Step S7: Screening critical and non-critical transmission sections using security risk indicators; Step S8: Searching for critical transmission sections in all operation scenarios and outputting the critical transmission section S of the power grid; This invention also relates to related systems. This invention can accurately and efficiently search for critical transmission sections of the power grid.

The application discloses a kind of city area photovoltaic cluster dynamic grid division methods based on space-time coupling tensor, comprising: synchronously collecting the active power time series and geographic coordinates of all photovoltaic nodes in city area;Build the heterogeneous space adjacency matrix considering electrical topology constraint, calculate the static space correlation degree between any two photovoltaic nodes;Build the time-shift cross-correlation matrix that captures the moving characteristics of weather system, quantify the dynamic time correlation degree between nodes;Perform space-time feature tensor fusion, obtain global space-time affinity matrix;Photovoltaic nodes are divided using improved spectral clustering algorithm, and the optimal grid number is adaptively determined by contour coefficient, finally generate photovoltaic cluster grid division scheme at this moment;According to grid drift threshold, trigger grid reconstruction.The application can effectively consider physical topology constraint and meteorological time delay characteristics, realize the accurate aggregation of photovoltaic cluster, significantly improve the accommodation capacity and regulation flexibility of distribution network to distributed energy.

The application discloses a kind of wind turbine power intelligent prediction method and device of multi-model synergic fusion, belong to wind turbine operation control and new energy prediction technical field.The method includes obtaining wind turbine data, wind turbine data is divided into several wind processes according to fixed sample length, each wind process index is calculated, and effective wind process set is obtained according to each wind process index;Obtain the feature vector of each effective wind process, obtain the similarity matrix using the feature vector, and divide each effective wind process into several working condition clusters based on spectral clustering algorithm;Each working condition cluster sub-model is constructed to each working condition cluster, and working condition cluster sub-model stack is formed;Global prediction model is constructed based on all effective wind processes, and global prediction model stack is formed;For the target wind process to be predicted, working condition cluster sub-model prediction value and global prediction model prediction value are obtained using working condition cluster sub-model and global prediction model, and power prediction value is obtained using dynamic weighting mode.

The invention relates to the field of remote sensing data classification, and discloses a multi-modal remote sensing data clustering method and device and a medium, and the method comprises the steps: obtaining multi-modal remote sensing data, and generating a superpixel segmentation result S through an entropy rate superpixel segmentation algorithm; adopting a k-nearest neighbor algorithm to initialize a graph structure of the multi-modal remote sensing data for the segmentation result S; filtering the graph structure to obtain an optimized graph structure; fusing the optimized graph structures to obtain a fused graph, and encoding the fused graph into a population in an evolutionary algorithm; using a multi-objective evolutionary algorithm to obtain the fusion image, and solving a corresponding optimization equation to obtain an optimal fusion image G *; a spectral clustering algorithm is applied to the optimal fusion image G *, and target image classification is completed; the method has the beneficial effects that not only is the calculation complexity effectively reduced, but also the optimization efficiency of the graph structure is improved; in addition, a globally optimal solution is realized by using a multi-objective evolutionary algorithm.

The invention discloses a machine learning spectral clustering analysis method and system for fatty liver auxiliary screening. Comprising the following steps: acquiring medical data and a fatty liver diagnosis result, and preprocessing and standardizing the medical data to obtain a training data set; reversely searching an optimal feature combination from a plurality of data indexes through an enumeration method; training the fatty liver auxiliary analysis model based on the spectral clustering algorithm by using the training data only containing the optimal feature combination; and obtaining a fatty liver auxiliary analysis result by using the trained fatty liver auxiliary analysis model. The method is used for early-stage rapid screening of the fatty liver diseases, the classification accuracy and auxiliary diagnosis efficiency of the fatty liver diseases are improved, and high calculation cost is not needed.

The invention provides a multimedia data security measurement method based on multi-dimensional association, and the method comprises the steps: dividing a plurality of features of a single image in multimedia data into a plurality of clustering groups based on spatial similarity through a spectral clustering algorithm for the multimedia data; and judging whether the multimedia data is safe or not by combining the content safety score, the privacy information safety score, the semantic safety score and the formalized safety score, thereby realizing optimization of safety evaluation and calculation efficiency.

The invention provides a historical building group temporary support system collaborative optimization method and system, and relates to the technical field of building protection, and the method comprises the steps: obtaining point cloud data through three-dimensional laser scanning, carrying out the component recognition and force transmission sequence analysis, carrying out the support partitioning through a spectral clustering algorithm, and carrying out the building group temporary support system collaborative optimization. And a group action strength matrix is constructed based on the geometric similarity and the structural coupling degree, and support parameters are optimized by using a genetic algorithm. The overall cooperative arrangement of the historical building group supporting system is achieved, the supporting efficiency and safety are improved, and the supporting cost is reduced.

The invention provides a multi-view subspace clustering cancer subtype identification method based on self-reinforcement learning, and the method comprises the steps: firstly, extracting the potential feature representation of each view from multi-omics data through a potential feature learning module; then, clustering similar samples by using a self-expression learning module, and introducing initial graph information as a supervision signal to construct a self-expression coefficient matrix; secondly, inputting the matrix into a view image fusion unit, and fusing multi-view information to generate a consensus image; in addition, in order to further suppress noise interference in multi-omics data, a self-strengthening back propagation unit is introduced, a confidence matrix is generated by optimizing a self-expression coefficient, fusion loss back propagation is guided, the quality of the self-expression coefficient is iteratively improved, and a consensus graph is optimized; and finally, based on the optimized consensus graph, realizing cancer subtype identification by applying a spectral clustering algorithm. According to the method, the self-reinforcement learning strategy is introduced, the interference of noise on sample relation capture is effectively relieved, and the clustering performance is remarkably improved.

The present application relates to the technical field of power system metering, in particular to a method and system for operation error diagnosis and analysis of electric energy meter based on electric energy meter operation error, wherein the method comprises: obtaining electric energy meter measurement value curve data of a target area in a set period from a metering automation system, and generating standardized electric energy meter operation data sequence through preprocessing; using an improved spectral clustering algorithm to carry out curve shape clustering on the data sequence, combining elbow rule to determine the optimal classification number of electric energy meter operation state, and completing the classification of electric energy meters in different operation states; relying on the principle of energy conservation to build a state evaluation model, screening effective data points through a data optimization strategy, reducing the model ill-conditioning by means of an improved Tikhonov regularization method, solving the error estimation equation to obtain the electric energy meter operation error estimation value and complete the state diagnosis. This method is suitable for processing massive electric energy meter time series data, smoothly weakening model operation interference, regularizing the electric energy meter operation error solving process, and optimizing the working condition diagnosis adaptability.

The present application relates to a kind of multi-center medical image lesion typing methods based on federal spectrum clustering, belong to medical image analysis technical field, including the following steps: S1: multiple medical institutions are regarded as federal learning client, and the medical image data and supporting clinical data of each client local are obtained and stored in each client local;S2: feature extraction and multimodal feature fusion are carried out in each client local, and multimodal lesion feature set is obtained;S3: center server and each client pass through iteration communication, and the multimodal lesion feature set is jointly clustered using federal spectrum clustering algorithm, until global clustering center converges;Only transmission encrypted statistical parameters in communication process;S4: according to clustering result, output lesion typing information.The present application solves the pain points of difficult data sharing and high privacy leakage risk in traditional centralized lesion analysis, and improves the lesion typing accuracy in heterogeneous data scenario.