59 results about "Dynamic clustering" patented technology

This invention discloses a comprehensive management method and system for base station power consumption, belonging to the field of base station power consumption. The method includes: topological mapping of the access relationships of AC / DC equipment within the base station to form a hierarchical power consumption association list; dynamic clustering parsing of the list to generate a time-series evolution chain reflecting state changes and remote control history; constructing a device differentiation identification model based on the time-series evolution chain, determining abnormal metering factors and outputting candidate correction events; labeling the corresponding areas of the candidate events as local anomalies, and generating a correction reference group by fusing multi-source information such as meter readings and sensor data; dynamically iteratively correcting the global metering results based on the correction reference group, and generating intelligent management instructions by combining the corrected load sharing ratio and branch authorization strategy. The system includes corresponding functional modules. This invention can systematically solve the problems of unclear base station power consumption topology, inaccurate metering, and extensive management, achieving high-precision metering and automated intelligent control.

This invention discloses a multi-parameter coupling-based deep peak-shaving full-condition optimization control method for coal-fired power units. Addressing the problems of existing technologies, this invention constructs a multi-parameter coupling index system, calculates the comprehensive coupling strength between each parameter and the load, and selects core parameters. Dynamic clustering is used to finely divide the full-condition operation from 20% to 100% rated load into multiple operating zones. An LSTM-GRU dual-channel deep learning prediction model is established for each operating zone. A three-layer hierarchical control architecture is designed, comprising load optimization scheduling, boiler-turbine coordination control, and combustion optimization execution. The boiler-turbine coordination layer uses multivariate model predictive control and online self-tuning of multi-objective weights through fuzzy inference. The combustion execution layer uses the NSGA-III multi-objective evolutionary algorithm to optimize air distribution. A constraint-adaptive tuning mechanism based on safety margin is established to form a closed-loop iterative optimization across all operating conditions. This invention can significantly improve the load response speed, main parameter stability, and combustion economy of the unit during wide-load operation.

The application discloses a wireless sensor network dynamic clustering routing method based on a gecko optimization, and belongs to the technical field of wireless sensor network routing optimization. The method first determines the target cluster head quantity in the current round based on the position, residual energy, neighbor relationship and survival state of a sensor node, constructs a candidate cluster head set and a cluster head selection fitness function; then represents the candidate cluster head node as a priority vector in a gecko optimization algorithm, determines the cluster head set in the current round through discretization mapping, iterative updating and fitness evaluation; then determines a data transmission path according to the communication cost between the cluster head node and a sink node, and completes clustering and data collection, fusion and forwarding; finally, performs local cluster head rotation in a cluster and global re-clustering according to the energy state of the cluster head node and the inter-cluster energy balance state. The method can improve the cluster head distribution balance, reduce network communication energy consumption and re-clustering control overhead, and prolong the network life cycle.

The application discloses a kind of alliance chain slice construction and consensus method based on dynamic clustering, comprising, the real-time computing power of alliance chain node, transaction frequency and transmission rate are collected;By verifiable random function, node is divided into Full_node of storing complete data and Slice_node of only storing slice data;DQN algorithm is used to determine the number of slices K, and the final consensus group is formed by selecting consensus representative nodes from Full_node based on dynamic performance threshold;The correlation matrix between the remaining nodes is constructed, and the nodes are evenly distributed to K slices by means of improved clustering algorithm;In each slice and consensus group, the Leader node is elected;When a new transaction arrives, execute the corresponding consensus process according to whether the transaction parties belong to the same slice, and write into the blockchain after verification, collect consensus performance data by Leader, and determine whether to trigger slice reconstruction according to performance data: if triggered, re-slice, otherwise continue to process transactions until completion;The application can improve the throughput of the overall system, relieve the data storage pressure, and improve the system response efficiency.

This disclosure provides a text data processing method, apparatus, device, storage medium, and program product, relating to the field of big data processing. The method includes: acquiring incremental text data and historical clustering data; the incremental text data includes at least one text data item; performing incremental clustering processing on the text data items based on historical clusters in the historical clustering data, matching historical clusters or creating new clusters for each text data item, and generating corresponding cluster topics for clusters that meet preset conditions, thereby obtaining incremental clustering results; and updating historical clustering data according to the incremental clustering results. This method can effectively integrate newly added text data, dynamically adjust the clustering results as data is updated, avoid the high overhead of full re-clustering, and enhance the interpretability of clustering results by automatically generating cluster topics. It achieves efficient dynamic clustering of incremental text data, improving the automation and intelligence level of data management.

The application discloses a kind of personalized federated continuous learning method and system based on dynamic clustering and multi-scale prototype, to solve the problem of spatiotemporal catastrophic forgetting in personalized federated continuous learning.The server initializes global model containing prompt parameters and cluster-level multi-scale prototype library, and the client generates routing histogram representing local data distribution and incremental prototype through a gated routing network and uploads it.The server calculates JS divergence based on routing histogram to detect drift, triggers K-Means online re-clustering, and dynamically updates the multi-scale prototype library containing short-term, long-term and drift prototypes in combination with the drift signal.The client uses long-term prototype to construct a joint loss optimization model.The application balances new knowledge learning and old knowledge retention while protecting data privacy and reducing overhead, and adapts to scenarios where data distribution continues to evolve.

This application provides an adaptive hierarchical distributed learning method and system for heterogeneous industrial production line data, relating to the field of distributed learning technology in industrial manufacturing. The method includes: constructing a federated learning framework for global training; uniformly aggregating low-level common visual features based on global model parameters to construct a shared feature space; performing mid-level adaptive clustering based on feature and gradient similarity to obtain a cluster-shared model; performing high-level fine-grained re-clustering and personalized optimization within clusters to construct personalized models; integrating model parameters from each layer to construct a tree-like multi-level aggregation structure and performing dynamic optimization; and finally ensuring training stability and generalization performance through cross-layer collaboration and convergence optimization. This application, through multi-level aggregation and dynamic clustering mechanisms, effectively solves problems such as semantic conflicts, low aggregation efficiency, and training instability caused by heterogeneous industrial production line data, significantly improving the model's convergence speed, stability, and generalization ability in cross-production line scenarios.

The application discloses a line intelligent planning and fault early warning method suitable for power engineering design and relates to the technical field of power engineering. The method comprises the following steps: through spatial semantic segmentation and a cost surface model, quantifying geographic information, geological conditions and construction cost into a continuous decision surface, combining a dynamic clustering algorithm to identify a large-area low-cost area as a primary feasible region; based on spatial features, geological stability and facility correlation data, constructing multi-dimensional decision indicators, and through real-time correlation coefficients, identifying "fault regions" such as terrain mutations or facility conflicts, and triggering an abnormal processing mechanism; selecting a seed region with the optimal construction condition as an anchor point, constructing a joint probability model of spatial features and cost, combining an expectation maximization algorithm and graph optimization technology, and generating an optimal path that takes into account the feasibility probability and cost benefit. The application improves the scientificity and economy of cable planning under complex geological conditions and provides key technical support for the intelligentization of power engineering.

The present disclosure belongs to the technical field of urban power grid safety control, and particularly relates to a source-load-storage modeling method, a power grid voltage control method and system. The method comprises: carrying out fine modeling of photovoltaic, energy storage and type-specific load, collecting equipment operation state data and identifying parameters; building a double-layer control architecture of centralized coordination-distributed collaboration, deploying centralized coordinators and distributed collaborators and defining functions; under the control architecture, measuring power distribution network topology, source and load real-time and historical data, and performing dynamic cluster division; based on historical optimal solution database initialization, using ADMM distributed optimization algorithm to calculate cluster total power adjustment of each cluster; based on cluster total power adjustment, voltage deviation factor and resource margin factor, combining the model and model parameters to perform power distribution of photovoltaic reactive power adjustment and energy storage active power adjustment on the equipment connected to the node. Due to fine modeling, the voltage control is provided with a basis, and the accuracy and reliability of the control can be improved.

The application discloses a grouping operation method, device and equipment of dense warehouse AGV and a storage medium, relates to the warehouse technology field, aims at the problem that the narrow channel of the dense warehouse leads to the operation channel conflict of multiple AGVs and the low efficiency of warehouse-in and warehouse-out, and provides a grouping operation method of dense warehouse AGV. The method collects the global AGV state and the warehouse-in and warehouse-out task queue through a sensor network, divides the AGVs into groups matched with the number of channels by using a position dynamic clustering algorithm, assigns the roles of goods putting and goods taking to each group by using a deep reinforcement learning decision model, issues task instructions by using an optimal assignment algorithm, monitors the operation state of the AGVs in real time and updates data, realizes the dynamic optimization of the scheduling strategy by model iteration, and ensures that only a single role group operates in the same channel. The application realizes the closed-loop automation of AGV scheduling, avoids channel conflict, and improves the collaborative operation efficiency and system turnover capacity of multiple AGVs of the dense warehouse.

The application relates to a single-phase ground fault early warning method and system based on dynamic clustering analysis, in the system, a data acquisition module collects real-time recording wave data of three-phase current and zero sequence current, and a characteristic waveform is constructed based on the recording wave data; a multi-scale feature extraction module calculates a plurality of dimensions of current feature vectors based on the characteristic waveform, normalizes the plurality of dimensions of current feature vectors, and generates a high-dimensional feature vector; an adaptive dynamic clustering module performs offline spectral clustering based on historical data to obtain an initial cluster; a local neighborhood graph is established with the initial cluster center as a node; a mode migration tracking module captures the dynamic change path of a new sample in a feature space based on the high-dimensional feature vector, calculates a mode migration recognition index and a Mahalanobis distance value; and a multi-level early warning decision module executes multi-level early warning determination based on the mode migration recognition index and the minimum Mahalanobis distance value, and generates corresponding early warning events when different early warning conditions are met.

The application relates to the technical field of wireless communication networks, and particularly discloses a cluster-based unmanned cluster ad hoc network system, which comprises a distributed spectrum situation awareness module, a dynamic clustering and cluster head election module, a spectrum-topology joint optimization routing module and a cross-layer cooperative control module. Through cooperative sensing of an electromagnetic environment, real-time spectrum quality is integrated into cluster head election and routing optimization, and closed-loop linkage and adaptive reconstruction of the three are realized, so that the spectrum utilization, link reliability and overall communication performance of the network in a dynamic environment are improved. The system introduces the distributed spectrum situation awareness module, realizes high-resolution and low-latency cooperative sensing of a complex dynamic electromagnetic environment, and provides real-time and accurate spectrum situation information for network decision-making.

The application discloses a mixed expert model word grouping scheduling method and electronic equipment, and relates to the technical field of artificial intelligence, which comprises the following steps: in response to the arrival of word units in the form of a stream to a preset buffer, obtaining the embedding vector features, routing history and arrival time corresponding to each word unit to splice into the target feature vector corresponding to each word unit, and the preset buffer is configured with a preset waiting time threshold and a capacity upper limit value; clustering and grouping each word unit according to the target feature vector to generate at least one task group; determining the priority of each task group based on the average waiting time of each word unit in each task group and the historical hot index; under the condition of meeting the limitation of the preset buffer, scheduling each task group according to the priority to enable the mixed expert model to perform operation on each word unit. The method of the application improves the utilization rate of experts and the system throughput while ensuring the upper bound of delay through short window buffering and dynamic clustering.

The present application relates to the technical field of intelligent quality inspection of the textile industry, and provides a silk flaw detection method based on machine vision, which comprises the following steps: preprocessing a gray image of a silk fabric surface and extracting a multi-fractal feature vector for storage in a dynamic cache pool; generating a flaw category cluster set through adaptive clustering analysis, and determining a typical defect sample based on the mean value of the cluster set; processing the sample through Gaussian-Laplacian filtering to generate a flaw morphology label, and creating a defect heat map sequence and a weaving time sequence stamp based on the same; generating a label-enhanced image by combining the heat map with the original warp and weft image of the fabric, and composing a loom operation cycle defect segment in time sequence; and finally sending the defect segment and the defect type code to a cloth inspection machine terminal. The method realizes efficient flaw detection through multi-scale feature analysis and dynamic clustering, and correlates the production time sequence for quality tracking. The present application can improve detection accuracy and equipment optimization efficiency.

The application provides an end-edge-cloud collaborative hierarchical federated learning training method, device and storage medium, and belongs to the technical field of distributed machine learning and edge computing. The method is applied to a three-layer system including a cloud server, an edge server and a terminal device, and comprises the following steps: the cloud server distributes a global model to each edge server; the edge server dynamically clusters according to the computing and communication capabilities of the subordinate terminal devices, determines differentiated model calculation compression rates and communication compression rates for different clusters, and distributes lightweight models; the terminal device performs local training and uploads parameters; the edge server and the cloud server perform teacher-free online knowledge distillation based on a shared data set to collaboratively update the model; and the cloud server triggers global aggregation according to a dynamic time threshold scheduling strategy to generate a new round of global model. The application effectively alleviates the training blocking problem in a heterogeneous device environment, improves resource utilization efficiency and model performance, and enhances system convergence stability.

This invention discloses a dynamic clustering method for space-based distributed collaborative detection. First, based on dynamically changing mission requirements and target status, a dynamic clustering algorithm optimized for collaborative detection performance is designed to cluster satellites within the constellation. Then, using the comprehensive capabilities of nodes as a crucial basis for node star election and cluster structure maintenance, a multi-factor weighted dynamic edge node star selection algorithm is designed, ultimately achieving a hierarchical, ordered, distributed collaborative architecture of "constellation-cluster-single star". This invention can reduce the information propagation cost within the constellation and improve information collaboration and mission execution efficiency.

This invention discloses a bearing fault diagnosis method based on a similarity clustering region transfer network, comprising: collecting vibration signals of known operating conditions and the bearing to be tested, and constructing source domain and target domain datasets respectively; selecting excellent samples in the source domain through scoring, and performing similarity assessment and statistical projection correction on the target domain samples according to the data distribution pattern of the source domain; inputting the two types of samples into a time-frequency domain feature extraction network to extract and fuse time-frequency domain features; constructing a dynamic clustering transfer module, combining dynamic entropy weight adjustment, Gaussian mixture model clustering and Hungarian algorithm to assign pseudo-labels, and iteratively training network parameters through a composite loss function; and realizing online fault diagnosis of the bearing to be tested after training. This invention effectively solves the problems of low diagnostic accuracy and poor versatility caused by differences in data distribution across operating conditions, improves the completeness of fault feature extraction and transfer reliability, and significantly improves diagnostic accuracy.

The application discloses a large model federated learning method and system based on low-rank subspace dynamic clustering. Each client of the application utilizes low-rank adaptation technology to perform efficient fine-tuning locally, and extracts a low-dimensional subspace of the parameter update quantity as a geometric representation through fast singular value decomposition; the server calculates the geometric similarity between the clients based on the subspace principal angle metric, constructs a bottom-up hierarchical clustering tree, and dynamically determines the optimal clustering structure by using the first-order difference of the link cost sequence; parameter aggregation is performed within the cluster to eliminate directional conflicts, and the intra-class aggregated model is used as the starting point for the next round of training, while a global model is introduced as a proximal constraint. While maintaining low communication bandwidth and computing cost, the application effectively aligns the optimization trajectory, eliminates the destructive interference of the gradient, and realizes the dual improvement of the generalization ability and personalized performance of the multi-modal large model in the heterogeneous data scene.

The present application relates to the technical field of concrete monitoring, in particular to a method for monitoring and predicting carbonation depth of concrete under dry-wet cycle conditions in salt-fresh water areas. The method comprises the following steps: by laying multiple source carbonation monitoring nodes, collecting carbonation related characteristic data, and using the maximum information coefficient algorithm to screen the dominant factor; using fuzzy C-means dynamic clustering algorithm to identify carbonation stage and spatial partition; constructing partition discriminant model through multiple discriminant analysis, and combining extreme learning machine to realize carbonation depth prediction and risk assessment, and finally generating carbonation risk early warning report. The present application realizes the whole process integration of multi-source data driving, feature screening, dynamic partition clustering, partition discrimination and intelligent prediction; improves the monitoring accuracy, risk identification rate and management response efficiency of concrete members, and can provide scientific and effective technical support for concrete durability evaluation and operation management in special environments such as ocean and estuary.

This invention relates to the field of distributed energy regulation in power distribution networks, specifically disclosing a photovoltaic cluster collaborative regulation system and method based on digital twins and topology association. The system includes a cloud-based decision-making layer, an edge control layer, and a terminal execution layer. The cloud deploys a global topology database and a digital twin engine to perform day-ahead optimization scheduling and generate regional power targets. The edge layer uses an improved NSGA-III algorithm incorporating topology constraint factors for multi-objective optimization, generating localized regulation commands. The terminal layer possesses four-quadrant enhanced control and communication interruption autonomy functions. The system constructs a three-level topology model of "distribution transformer-branch line-equipment," quantifies electrical distance and topology association, and establishes a dynamic adjacency matrix. This invention achieves topology-driven precise power allocation and dynamic clustering, effectively improving photovoltaic absorption capacity, regulation response speed, and power distribution network operation safety through digital twin pre-simulation and cloud-edge-terminal closed-loop collaboration.

The application discloses a kind of correlation analysis methods, systems, devices and media of alarm data, wherein the method obtains several multi-source alarm data;All the multi-source alarm data is extracted to multidimensional feature, and several multidimensional feature vectors are obtained;All the multidimensional feature vectors are dynamically clustered and compressed, and several alarm compression clusters are obtained;The alarm compression cluster includes several alarm cluster points, and the alarm compression cluster records the space-time distribution characteristics of all the alarm cluster points;All the alarm compression clusters are analyzed by graph structure correlation, and the alarm correlation analysis result of the multi-source alarm data is obtained.The method can effectively improve the correlation analysis efficiency and effect of alarm data.The application relates to the technical field of communication network operation and maintenance.

The application discloses a kind of based on dynamic clustering and agent's flexible fee control strategy optimization method and system, it is related to electric power service management technical field.The method includes the following steps: establishing flexible fee control recommendation agent according to user historical power consumption behavior data, and constructing fee control service professional sub-label system;According to the actual power consumption of user, formulate flexible fee control strategy, use clustering analysis algorithm to carry out user group to flexible fee control recommendation agent, and formulate flexible fee control strategy based on group differentiation;According to the flexible fee control recommendation agent, the power consumption characteristics of user are fused and analyzed and predicted using Guangming electric power big model, and the flexible fee control strategy based on user individualization is generated;According to the process optimization of flexible fee control strategy execution situation.The present application can generate differentiated and intelligent fee control scheme based on user electric power image.

The present application provides an open-pit mine network security situation awareness method, device and electronic equipment, relating to the technical field of network security. In view of the particularity of the open-pit mine network, combined with its characteristics such as multiple mobile nodes, complex industrial protocols and harsh physical environment, through multi-dimensional data fusion, dynamic clustering optimization, cross-modal large model reasoning and industrial honeypot trapping, etc. The present application realizes accurate perception and prediction of the open-pit mine network security situation. The present application reduces the dimension of open-pit mine network security situation awareness, improves the perception efficiency and accuracy in complex environment, and provides a strong guarantee for the safety of open-pit mine key infrastructure.

This invention relates to a method for improving the quality of equipment designation data based on multi-source heterogeneous data fusion and clustering, belonging to the field of data governance. This invention focuses on core pain points in equipment designation data governance, such as the difficulty in integrating multi-source heterogeneous data, poor adaptability of anomaly detection, and difficulty in handling semantic contradictions. It employs multi-source heterogeneous data fusion preprocessing, a dynamic clustering anomaly detection model, and domain knowledge-driven anomaly correction to achieve a breakthrough in the entire process. This invention can significantly improve the quality of equipment designation data, achieving industry-leading accuracy in key parameters and effectively reducing subsequent development risks and costs.

This invention relates to a method and system for identifying lithofacies in tight carbonate gas reservoirs based on dynamic clustering and bidirectional long short-term memory networks. Belonging to the field of oil and gas exploration and production technology, it addresses the problems of low accuracy in lithofacies identification results and insufficient generalization performance of existing identification models. The method includes: first, employing a standardized preprocessing method to reduce the interference of differences in logging parameter dimensions on the learning process; second, using a dynamic clustering algorithm to perform unsupervised classification of logging data, through similarity adaptive construction, dynamic determination of cluster numbers, and iterative update mechanisms, enabling the clustering results to more realistically reflect the complexity and heterogeneity of the internal lithofacies structure, achieving automatic lithofacies labeling; finally, constructing a BiLSTM-Attention model, introducing an attention mechanism to capture the temporal dependencies between logging curves, achieving intelligent lithofacies identification. This invention is applicable to lithofacies identification in tight carbonate reservoirs.

This invention provides a real-time early warning method for casing deformation during shale gas fracturing based on the spatiotemporal evolution characteristics of microseismic data. The method includes: real-time acquisition of microseismic data from the fracturing area; dynamic clustering of the microseismic data, reconstructing the clustered event set into an event cloud object with spatiotemporal continuity; calculation of morphological and dynamic parameter fields for the event cloud object; constructing a multi-dimensional temporal feature vector from the calculated morphological and dynamic parameters and construction pressure data, and inputting it into an early warning model; and generating graded early warning signals for casing deformation based on the output of the early warning model. The early warning signals include four levels: blue, yellow, orange, and red. This invention can capture precursors to fault activation, overcome the limitations of single parameters, achieve advanced real-time early warning, reduce the risk of casing deformation, and meet the needs of efficient shale gas development.

This invention relates to the field of farmland soil remediation technology, and in particular to a comprehensive evaluation method for the remediation effect of heavy metal-contaminated farmland soil. The method includes: constructing a four-criteria hierarchical evaluation system encompassing pollution load, ecotoxicity, soil health, and remediation cost; selecting at least three time points covering the initial, middle, and late stages within the remediation cycle, and simultaneously collecting index data for the divided remediation units; after standardizing the data, integrating subjective and objective weights using a combined weighting method based on the optimization of the sum of squared deviations; calculating a comprehensive evaluation index for each node, determining grade thresholds and classifying them using dynamic clustering; generating a spatiotemporal evolution map, identifying key limiting factors for "medium" and "poor" grade units, and outputting a diagnostic report containing dynamic evaluation, heterogeneity analysis, and optimization suggestions. This invention provides comprehensive evaluation, reliable data, and objective results, and can support precise control of remediation projects.