25 results about "Gray relational analysis" patented technology

The invention relates to a multi-element reliability test data dynamic weight optimization method, which comprises the following steps of: inputting data, including accelerated life test data, fault excitation data, a mean value of historical failure and a standard deviation; calculating entropy values and dynamic weights of various types of data according to a formula; according to the dynamic weight optimization method, weight distribution is performed on existing test data related to various reliability faults through a multi-source data weight distribution method fusing an entropy weight method and grey correlation analysis, so that a verification result is more credible. According to the entropy-gray combined calculation technology, the theory of information entropy and the theory of gray relational degree are integrated, and test data are adjusted through the product term of the information entropy and the gray relational degree to verify the distribution weight. According to the full-automatic weight distribution system, a real-time weight updating mechanism without manual intervention is achieved by updating an input sequence of data.

The invention discloses a cibotium barometz sporophyte transformation method and system based on multi-factor coupling regulation, and relates to the technical field of biology. The method comprises the following steps: synchronously acquiring environment data and sporophyte image data; preprocessing the environment data, performing feature extraction on the image data, and splicing the processed data into a comprehensive feature vector; quantifying the weight, correlation degree and interaction level among the environmental factors through an entropy method, grey correlation analysis and a coupling coordination degree model; constructing an improved CNN-LSTM neural network, extracting features through a double-branch structure, fusing a coupling analysis result to carry out cross-modal fusion, and training to obtain a prediction model; when the predicted conversion rate is lower than a threshold value, environment parameters are automatically adjusted until conversion is completed; according to the method, the problems of environmental factor data isolation and regulation lag in the prior art are solved, accurate prediction and intelligent regulation of the sporophyte transformation process are realized, and the transformation success rate and the production efficiency are remarkably improved.

The invention discloses an electricity purchasing and selling strategy optimization aided decision-making method, and the method comprises the steps: determining an evaluation index set, and evaluating the gray correlation degree of each evaluation index through gray correlation analysis. Based on the grey correlation degree, the weight vector of the evaluation index is further determined so as to comprehensively consider multiple factors influencing the electricity market, such as market transaction share, electrical load and generating capacity, and meanwhile, the influence degree of each index on the electricity market is more accurately evaluated. Then, a fuzzy evaluation matrix of the evaluation indexes is constructed, and a fuzzy comprehensive evaluation vector is calculated in combination with the weight vectors of the evaluation indexes; the vector can comprehensively reflect the comprehensive membership degree of a plurality of evaluation indexes to high, medium and low risk levels, so that comprehensive evaluation of market transaction risks is realized. In the decision-making process, the particle swarm optimization algorithm is adopted to solve the electricity purchasing and selling strategy decision-making model, the optimal electricity purchasing and selling strategy is automatically explored, and the decision-making efficiency and accuracy are remarkably improved.

This invention discloses a health status prediction system and method for electric shovel equipment based on edge-cloud collaboration and digital twins, belonging to the field of electric shovel equipment status prediction technology. The system includes an electric shovel equipment health status management system based on edge-cloud collaboration and an electric shovel digital twin model. It collects multi-dimensional status data of the electric shovel through multiple sensors, and achieves layered transmission and edge-cloud collaborative processing via NB-IoT, edge devices, and 5G base stations. A high-fidelity electric shovel digital twin model is constructed to achieve synchronous mapping between the physical entity and the virtual model. The method utilizes grey relational analysis to calculate the correlation between real-time status data and offline fault samples, accurately assessing the equipment's health status, and outputs predictive maintenance processes through edge-cloud collaboration, synchronously updating the cloud database. This invention achieves closed-loop management of electric shovel equipment from data acquisition, twin modeling, fault diagnosis to predictive maintenance, significantly improving the accuracy of status assessment and the efficiency of operation and maintenance collaboration.

This invention discloses a method, apparatus, equipment, and medium for information integration processing based on power grid fault analysis, relating to the field of information integration processing technology for power grid fault analysis. The method includes: constructing a multi-source fault analysis model of the power grid based on pre-acquired multi-source fault feature data using a grey relational analysis algorithm; generating a data feature sequence based on the multi-source fault analysis model using a wavelet transform algorithm; generating a time-scaled calibration result based on the data feature sequence using a dynamic time warping algorithm; updating the multi-source fault feature data using a time-series interpolation algorithm based on the time-scaled calibration result to obtain a corrected fault feature sequence; and fusing the acquired power grid topology and the corrected fault feature sequence using a support vector machine algorithm to generate a fault judgment result. This application aims to solve the problems of low time-scaled calibration accuracy and lack of topology correlation in traditional methods.

This invention discloses an early detection and evaluation method and system for the abiotic stress tolerance of new forest tree germplasm. The method includes: collecting chlorophyll fluorescence indicators, phenotypic indicators, and physiological and biochemical indicators of forest trees under abiotic stress conditions, and performing data cleaning and standardization; performing principal component analysis on the multi-source standardized dataset to screen candidate indicators; performing grey relational analysis using the comprehensive score of forest tree stress tolerance as a reference sequence and the candidate indicators as a comparison sequence; determining the core indicators and their weights based on the variance contribution rate of the principal component analysis and the correlation degree of the grey relational analysis; obtaining the abiotic stress assessment index from the core indicators using a membership function; and classifying the tolerance level of forest tree germplasm using a clustering algorithm based on this index. This invention achieves early, rapid, and accurate evaluation of the abiotic stress tolerance of forest tree germplasm, and can effectively guide forest tree stress resistance breeding work.

This invention discloses a method for constructing a time-series prediction model for the water environment based on dynamic feature mining of data, belonging to the field of environmental engineering technology. It solves the problems of high complexity and low efficiency in existing methods for dynamic feature mining of water environment data. The invention identifies target variables for a target watershed, collects historical data of the target variables and related variables, extracts features from the historical data, and establishes an initial input dataset. It uses grey relational analysis to analyze the correlation between the features of the target variables and the features of related variables, filtering variables whose correlation with the target variables exceeds a threshold. It uses SSA to optimize the parameters of STL, and uses the optimized STL to decompose the historical data of the target variables. Finally, it establishes an LSTM model, trains the LSTM model using the decomposed data and variables whose correlation exceeds the threshold, and obtains the SSA-LSTM model. This invention is applicable to water environment prediction and dynamic feature mining.

This invention discloses a comprehensive evaluation method for multi-dimensional dynamic equilibrium of flexible operation of coal-fired power units, comprising: determining a comprehensive evaluation index system for coal-fired power units and calculating the values ​​of each tertiary evaluation index; calculating the objective weights of secondary evaluation indicators using the entropy weight method, and introducing a dynamic group decision-making method based on the analytic hierarchy process to obtain a subjective dynamic group decision-making model, thereby determining the dynamic subjective weights of the secondary evaluation indicators; calculating the dynamic subjective-objective combined weights of each secondary evaluation indicator by combining and weighting the subjective and objective weights; and calculating the closeness between the primary evaluation indicators and the positive and negative ideal solutions using the approximation ideal solution ranking method and the grey relational analysis method based on the dynamic subjective-objective combined weights of the secondary evaluation indicators, as the final evaluation result of the primary evaluation indicators, thereby achieving a multi-dimensional dynamic equilibrium comprehensive evaluation of the operating status of coal-fired power units at the second-level scale. This invention can realize a multi-dimensional dynamic equilibrium comprehensive evaluation of coal-fired power units at the second-level scale during flexible operation with deep peak shaving and frequency regulation.

This invention pertains to the field of intelligent technology in the catering industry, specifically a training method for a large-scale reasoning language model based on task vector fusion. It involves collecting comprehensive business data from the catering industry, achieving cross-modal entity alignment based on fractional-order grey relational analysis, constructing a domain knowledge graph including a causal relationship layer, and enhancing the corpus. A domain-adaptive base model is obtained through pre-training with causal language modeling loss as the primary loss and combined with dual-auxiliary loss. Task vectors are then fine-tuned and purified through adaptive low-rank decomposition. Adaptive task weights are calculated based on Sobol global sensitivity analysis, and task vectors are nonlinearly fused using Caputo fractional derivatives to obtain a multi-task fusion model. Finally, multi-source causal evidence is fused based on Riemannian manifold geodesic causal distance and D-S evidence theory to enhance the model's causal reasoning training. This method improves the model's multi-task balance performance and causal reasoning accuracy.

This invention relates to the field of safety evaluation technology for highway bridge erecting machines, and discloses a method for safety evaluation of the initial installation state of a highway bridge erecting machine. First, a multimodal safety evaluation model is established, including a structural layer, a dynamic parameter layer, and a risk factor layer. Then, a fuzzy clustering algorithm is used to divide the feature intervals of the dynamic parameter layer, and a grey relational analysis method is used to calculate the correlation coefficient of risk factors. The objective weights of the parameters in the dynamic parameter layer are determined using the entropy weight method, and the two are combined to obtain a comprehensive weight. Next, multi-source sensor data of the bridge erecting machine installation are collected in real time and discretized. Finally, the risk index is calculated layer by layer to determine the safety level of the initial installation state of the bridge erecting machine. This method comprehensively considers the structure, dynamic parameters, and risk factors of the bridge erecting machine, uses multiple algorithms to improve the accuracy of the evaluation, and achieves automatic determination of the safety level. It can effectively reduce the installation risk of the bridge erecting machine and ensure construction safety and the smooth progress of the project.

The invention relates to a layered distillation method for botnet detection in encrypted DNS (Domain Name Server) traffic, which mainly comprises the following steps of: carrying out adaptive feature extraction and screening on preprocessed data by combining an entropy weight method and grey correlation analysis, and screening out optimal features which are strongly correlated with botnet behaviors; classifying the botnet by adopting a spatial clustering classification strategy to obtain a plurality of different types of botnet features; a hierarchical knowledge distillation model HLD-TCN is constructed and trained; deploying a trained multi-student model; and inputting the real-time encrypted DNS traffic, sequentially carrying out parallel detection on the real-time encrypted DNS traffic through the student models, and if a certain student model is judged to be abnormal, immediately marking the real-time encrypted DNS traffic as Botnet traffic and terminating subsequent detection. The method has the advantages that the accuracy of botnet detection in the encrypted DNS traffic is improved, and an efficient and lightweight solution is provided for network security protection.

This invention discloses a project portfolio optimization method based on heterogeneous networks, relating to the fields of project management and intelligent decision-making. The method first acquires project objectives, capability requirements, resource inputs, task dependencies, and risk information, and then standardizes and encodes projects, capabilities, tasks, resources, and strategies. Based on RDF triples, it constructs project association networks, capability association networks, and capability-project fusion heterogeneous networks. It uses grey relational analysis and entropy weighting to determine capability importance, and combines project support, project contribution, data envelopment analysis, CRITIC, critical path method, and TOPSIS to complete project value, core projects, critical path, and risk assessment. Under budget, capability thresholds, project mutual exclusion, and schedule constraints, a multi-objective portfolio optimization model is established, and NSGA-II is used to generate a Pareto optimal solution set. Finally, a recommended solution is output through weighted scoring. This invention can improve the systematicity, accuracy, and risk controllability of resource allocation.

This application provides a method and system for analyzing and modeling user electricity consumption data based on semi-supervised learning, belonging to the field of power system data analysis and prediction technology. It solves the problems of low data verification accuracy and large deviations in characteristic correlation in existing technologies. The method includes the following steps: based on user electricity consumption characteristic data and climate characteristic data, a semi-supervised learning-driven electricity consumption data verification and tracking feedback model is constructed using a graph convolutional network; the Pearson correlation coefficient between electricity consumption characteristics of different types of users is calculated, and a characteristic correlation model based on a deep belief network is constructed to analyze the nonlinear relationship between user type, user electricity consumption time-series characteristics, and climate characteristics; based on grey relational analysis, the distribution characteristics of electricity consumption changes are captured. This application is applicable to power system resource optimization, especially improving grid operation efficiency under climate variability.

The invention discloses a coal-fired unit flexible operation dynamic comprehensive evaluation method and system with adaptive, robust and multi-criterion evaluation, and the method comprises the steps: constructing an evaluation index set through employing a fuzzy Delphi method; based on Kalman filtering dynamic fusion, using a subjective weight calculated by a triangular fuzzy number best-worst method and an objective weight calculated by an entropy weight method to obtain a dynamic subjective and objective combination weight of the multi-dimensional index; a comprehensive evaluation result is obtained by adopting optimal preference coefficients of a Monte Carlo simulation optimization approximation ideal solution sorting method and a grey correlation analysis method, and multi-dimensional dynamic comprehensive evaluation of the wide-working-condition operation state of the coal-fired unit is achieved; two new indexes of the time ranking fluctuation entropy and the cumulative ranking offset area are provided for evaluating the robustness of the comprehensive evaluation method, and it is ensured that the evaluation method is scientific and reliable. According to the method, an effective tool is provided for scientifically evaluating the wide-load operation performance of the coal-fired unit and optimizing the flexible adjustment capability of the coal-fired unit, and the competitiveness of coal power enterprises in the electric power spot market can be improved.

This invention provides a blockchain asynchronous consensus method based on GRA-HB-BFT, belonging to the field of blockchain asynchronous consensus technology. This invention employs grey relational analysis combined with entropy method to construct a dynamic node reputation evaluation model, objectively evaluating consensus node performance through multi-dimensional indicators and outputting a comprehensive reputation value. Secondly, the traditional asynchronous common subset protocol is decoupled into a reliable broadcast protocol area and an asynchronous Byzantine protocol area, enabling parallel execution of the two areas. Consensus nodes perform threshold encryption on transaction proposals and broadcast them. Then, based on the reputation evaluation results, broadcast instances of high-reputation consensus nodes are selected as input for Byzantine consensus, effectively reducing communication overhead. Finally, after parallel processing, the encrypted proposals are threshold decrypted, verified, and sorted before being submitted to the ledger. Simultaneously, the comprehensive reputation value is dynamically adjusted based on the consensus node performance, forming a continuously optimized closed-loop control mechanism.

The application discloses a discrete element contact parameter prediction method based on feature analysis-machine learning, comprising the following steps: selecting a discrete element linear parallel bond model for contact parameter calibration according to soft rock particle characteristics and mold material parameters, and determining model parameters and value ranges; obtaining contact parameters of the discrete element linear parallel bond model based on a small amount of uniaxial compression and Brazilian splitting tests, and constituting a contact parameter-soft rock macro feature low-fidelity dataset based on the contact parameters; determining key contact parameters affecting the soft rock macro feature based on grey relational analysis (GRA), Pearson correlation analysis and mutual information (MI); obtaining a contact parameter-macro feature high-fidelity dataset based on a large amount of uniaxial compression and Brazilian splitting tests; establishing a key contact parameter prediction model of the discrete element linear parallel bond model based on machine learning; and determining an optimal prediction model of the key contact parameters of the linear parallel bond model. Corresponding systems, electronic devices and computer readable storage media are also disclosed.

This invention discloses a multi-faceted and refined method for predicting electric vehicle charging load, relating to the field of load prediction technology. The method includes the following steps: Step S1: Constructing a system of influencing factors for electric vehicle charging load and quantifying the time series of each influencing factor; Step S2: Based on the system of influencing factors for electric vehicle charging load, using the information gain method to calculate the correlation between different influencing factors and electric vehicle charging load, and filtering them according to a descending order principle to determine the input factor type; Step S3: Selecting similar days for the electric vehicle charging load of the predicted day using grey relational analysis and DTW distance analysis to determine the input order of the input factors; Step S4: Building a prediction model using a deep belief network, and using the determined input factor type and the input order of the input factors as the input to the prediction model, outputting an effective prediction of the electric vehicle charging load.

The invention provides a small watershed driving-obstacle double-diagnosis key index identification method and system, and belongs to the technical field of hydraulic engineering based on computer data processing. The method comprises the following steps: firstly, selecting a target small watershed, analyzing water and sediment-ecological-economic system indexes of the small watershed, and constructing a small watershed basic data set; secondly, a key driving factor is obtained through grey correlation analysis; then, acquiring key obstacle factors by using an obstacle degree model; and finally, integrating the driving factor and the obstacle factor, determining a key influence factor of the target small watershed, and visually outputting a diagnosis result. The key index identification method based on driving-obstacle double diagnosis has the advantages of being comprehensive in diagnosis, high in applicability and high in decision support, and can provide scientific basis for small watershed water resource management, ecological protection and economic sustainable development.

The invention relates to the technical field of grain safety, in particular to a grain industry chain supply chain toughness evaluation method, which comprises the following steps: S1, collecting original data of a plurality of samples, each sample comprising a plurality of evaluation indexes; s2, calculating the weight of each evaluation index in each sample by adopting a panel entropy weight method; s3, constructing a TOPSIS-grey correlation analysis integration model, analyzing the close degree of each evaluation index and the optimal solution, and determining the sorting priority of each evaluation index; and S4, constructing an obstacle degree model, and measuring the obstacle degree of each evaluation index. According to the method, the panel entropy weight method is adopted to determine the weight, the TOPSIS-grey correlation analysis method is combined to comprehensively evaluate the toughness level of the supply chain of the grain industry chain, and key factors and weak links restricting toughness improvement are identified by means of the obstacle degree model, so that reference is provided for improving the toughness evaluation of the supply chain of the grain industry chain in China.

The invention relates to the technical field of multi-project screening, and discloses a gray correlation analysis-based complex equipment development multi-project screening method and system, and the method comprises the steps: carrying out the design of a complex equipment development multi-project selection room frame through a quality room, and constructing a complex equipment development multi-project sorting model; and considering the index data interval value condition of the complex equipment development multi-project, and constructing a multi-source uncertain complex equipment development multi-project sorting model to determine a project sequence. According to the method, a complex equipment development multi-project selection room is established, a complex equipment development multi-project sorting model is established, a good decision reference is provided for a complex equipment development enterprise in scientific research project establishment and selection, and the method can also be applied to evaluation of complex equipment development projects and determination of importance degree sorting of alternative projects, so that the research and development efficiency of the complex equipment development enterprises is improved. And the development project is preferentially selected, so that the method has a relatively high application value for the development work of complex equipment projects.

The invention discloses a user behavior model construction and classification evaluation method based on improved grey correlation, and relates to the technical field of electric power big data analysis, and the method comprises the steps: collecting the historical behavior data of an electric power user, and carrying out the preprocessing of the collected historical data of the user; according to the method, historical data of a user is processed by using a dynamic trend grey correlation analysis algorithm, and a grey correlation feature set reflecting an internal change rule of a user behavior is constructed by calculating a trend correlation degree taking amount, time, usage and number of people as central vectors; an incremental FP-Growth algorithm is adopted to mine data fusing original features and gray correlation features, frequent patterns and strong correlation rules in user behaviors are dynamically found and updated, finally, a quantitative model with counting times, frequency and duration as indexes is established based on a mined rule set, a user behavior testing and writing model is constructed, and a user behavior testing and writing model is established. And thus, a classified user behavior evaluation system composed of multiple dimensions is formed, and accurate portraits of the power users are realized.

This invention discloses an enterprise-level SSD resource scheduling system based on storage virtualization, relating to the fields of storage scheduling and data processing technology. It includes a virtual resource modeling module that performs aggregation and association processing, service level binding processing, and access feature organization processing on tenant identifiers, service volume identifiers, namespace identifiers, virtual function identifiers, IO queue occupancy information, and access request characteristics to generate virtual resource status data. In this invention, the virtual resource modeling module performs aggregation and association, service level binding, and access feature organization; the media status awareness module performs disk-level correspondence organization, time-sequence organization, and media pressure merging; and the dual-state mapping construction module performs resource instance correspondence processing, media bearer association processing, and gray relational analysis algorithm processing based on dynamic time warping alignment. This achieves continuous mapping between virtual resource instances and media operating states, explicit bearer relationships, and unified cross-layer state expression.

The invention provides a block chain asynchronous consensus method based on GRA-HB-BFT, and relates to the technical field of block chain asynchronous consensus, grey correlation analysis is combined with an entropy evaluation method to construct a dynamic node reputation evaluation model, the performance of a consensus node is objectively evaluated through a multi-dimensional index, and a comprehensive reputation value is output; secondly, a traditional asynchronous common subset protocol is decoupled into a reliable broadcast protocol area and an asynchronous Byzantine protocol area, parallel execution of the two areas is achieved, and a consensus node carries out threshold encryption on a transaction proposal and broadcasts the transaction proposal; then, based on a reputation evaluation result, a broadcast instance of a high-reputation consensus node is preferably selected as the input of Byzantine consensus, so that the communication overhead is effectively reduced; and finally, after parallel processing is completed, performing threshold decryption, verification and sorting on the encrypted proposal, submitting the encrypted proposal to an account book, and dynamically adjusting a comprehensive reputation value according to consensus node expression to form a continuously optimized closed-loop control mechanism.

This invention discloses a method for predicting sand production in oil wells based on production dynamics and ESP (Electric Submersible Pump) operating parameters, comprising the following steps: S1: a primary sand production factor correlation analysis strategy based on oil well production dynamic parameters; S2: a secondary sand production pattern prediction strategy based on ESP pump current. On one hand, this invention constructs a multi-dimensional dynamic parameter evaluation system by integrating grey relational analysis and analytic hierarchy process (AHP), establishing a quantitative correlation model between production parameter fluctuations and sand production levels, and realizing dynamic screening and weight allocation of sensitive factors of operating parameters. On the other hand, based on the temporal feature capture capability of LSTM networks, a two-dimensional feature space of pump current change rate and amplitude is constructed. By deeply mining the coupling relationship between current waveform and sand particle transport, a dynamically adaptive sand production state identification mechanism is established.

This invention relates to a carbon emission prediction method and system based on meta-learning data augmentation. The method acquires and preprocesses historical carbon emission time-series data and multi-dimensional influencing factors for a target region to construct an original dataset; it then constructs a few-shot time-series generation model and introduces time-series datasets from other domains for cross-domain meta-learning pre-training, fine-tuning it in the target domain to generate synthetic data, which, together with the original data, forms an augmented dataset; next, it filters key influencing factors through improved grey relational analysis and a lightweight attention mechanism; finally, using the augmented data and key influencing factors as input, it constructs a Transformer-based prediction model and outputs the predicted future carbon emissions for the target region. Compared with existing technologies, this invention has advantages such as improved accuracy, stability, and robustness in carbon emission prediction.