49 results about "Grey clustering" patented technology

The invention discloses a network security situation prediction method in the technical field of network information security. The method comprises the following steps of: analyzing the hazard degrees of each network security threat by using a grey clustering analysis method, and further constructing a hierarchical network security situation index system, obtaining network security situation values of each time monitoring point, constructing a time sequence, constructing the time sequence into a training sample set, and performing iterative training on the training sample set by utilizing an integrated learning Boosting algorithm to obtain a weak learning machine sequence meeting error requirements; obtaining a strong learning machine by utilizing the method for calculating the weighted sum of the weak learning machine sequence; and finishing predicting the network security situation values of future time monitoring points by using the strong learning machine. The method is relativelyhigher in adaptability and relatively lower in prediction error rate in terms of reduction in network security situation value prediction errors.

The invention provides a method and a device for online evaluation of a converter valve state based on grey clustering. The method comprises: acquiring online monitoring data of a converter valve in apreset time period; establishing a state evaluation indicator matrix of the converter valve according to online monitoring data, and calculating a score value of each evaluation parameter class in the state evaluation indicator matrix according to the online monitoring data; respectively constructing a whitening weight function corresponding to each preset gray class according to at least three preset grey classes; according to a preset weight value of each evaluation parameter class and the whitening weight function of the score value of each evaluation parameter class, using a grey clustering algorithm to respectively calculate a clustering weight value of each preset grey class; and evaluating an operating state of the converter valve according to the each clustering weight value. By implementing the method, real-time evaluation of the operating state of the converter valve is realized without manual operation, accuracy of an evaluation result is improved, and potential risk prediction of the converter valve is realized, and provides good data basis for planned maintenance of the converter valve.

The invention relates to the technical field of petroleum well drilling, in particular to a rapid drill bit optimal selection method based on foundation pit anti-drill parameters. The rapid drill bitoptimal selection method comprises the step of stratum sensitive anti-drill parameter extraction, the step of drill bit type selection statistics mode establishing and the step of drill bit optimal selection gray clustering, and the rapid drill bit optimal selection method is a rapid drill bit optimal selection method in which according to various well logging data of a well area where a to-be-drilled well is located, stratum sensitive anti-drill parameters are simply and accurately extracted, the drill bit type with the fast footage for different stratum well depths is selected through comparison, a stratum anti-drill parameter and corresponding efficient drill bit type big data mode base is established, a gray dynamic clustering analysis method is adopted, and the optimal drill bit typeis given for different well section stratums of a drilled well or a drilling well or the to-be-drilled well.

The invention discloses an airborne network safety risk assessment method based on improved AHP (Analytic Hierarchy Process)-GCM. The airborne network safety risk assessment method based on improved AHP-GCM includes the following steps: 1) analyzing a novel airborne network architecture risk source; 2) determining an assessment index system for the risk source and the threat state of the risk source respectively; 3) assigning the value to an index weight and examining the consistency; 4) solving the index weight of each layer and performing normalization processing; 5) dividing the risk levels, that is, determining the grey class; 6) establishing a white function; 7) calculating the grey clustering coefficient and determining the belonged grey class; and 8) based on the above steps, performing secondary assessment on the threat state of the risk source. The airborne network safety risk assessment method based on improved AHP-GCM determines the risk source and the risk state, on the basis of analyzing the novel airborne network architecture, assesses the risk source, on the basis of the analytic hierarchy process and the grey clustering method, performs secondary assessment on the threat state of the risk source, and integrates two times of assessment to optimize the risk assessment result to prevent bias and inaccuracy of single time of assessment result.

The invention discloses a science and technology service quality evaluation method and device based on combined weighting and fuzzy grey clustering. The method comprises the steps: analyzing the service quality of a science and technology resource service platform, obtaining an evaluation result of the service quality according to a constructed index set, testing index data, calculating the subjective weight and the objective weight of the index data according to a majority set weighted analytic hierarchy process and a weighting method, combining the subjective weight and the objective weight by adopting a combined weighting method, substituting an evaluation object into a whitening weight function by utilizing a grey clustering theory to calculate whitening values contained in different clustering indexes, obtaining a secondary index service quality evaluation result, and performing primary index evaluation according to an evaluation index system and the secondary index service quality evaluation result to complete evaluation of the service quality of the science and technology resource service platform. According to the invention, the model can obtain an accurate and comprehensive service quality evaluation result according to limited sample data, and has guiding significance for improving the science and technology service level and promoting opening and sharing of science and technology resources.

The invention discloses a construction behavior safety risk analysis and dangerous point identification method based on a knowledge graph. The method comprises the following steps: (1) establishing a construction safety accident knowledge graph according to a related construction safety accident report; (2) acquiring accident occurrence data according to the construction safety accident knowledge graph, analyzing field behavior safety data through a modified grey clustering algorithm, and calculating a construction field safety risk level; (3) calculating a behavior index correlation degree through a grey correlation analysis method, and tracking a key behavior index; and (4) identifying dangerous positions related to the key behavior indexes through a knowledge graph path reasoning algorithm. According to the method, objective data of a construction safety accident report are integrated by introducing the knowledge graph, risk feature data are extracted from the knowledge graph, a risk analysis algorithm is improved, dangerous point identification is supported, and the problem that an existing risk analysis and dangerous point identification method is difficult to reflect project and regional features and subjective uncertainty is solved.

The invention relates to a power distribution network repair ability assessment method based on improved grey clustering. The method comprises the steps that S1, the analytic hierarchy process is adopted to establish an assessment indicator system; S2, original indicator values of all electric power companies are collected to obtain an original indicator matrix, and the original indicator values are subjected to dimension removal to obtain a standard indicator matrix; S3, the original indicator matrix is normalized, and weights are established for indicators according to the information entropy theory to obtain an entropy weight matrix; S4, a grey clustering method is adopted to set the number of grey clusters and a whitening weight function, and a saturability matrix corresponding to each grey cluster is calculated; S5, fuzzification operation is performed according to the saturability matrix obtained in the step S4 and the entropy weight matrix to obtain a clustering assessed value matrix, and the grey clusters which all the electric power companies belong to are determined according to the maximum value of elements in each row of the clustering assessed value matrix. Compared with the prior art, power distribution network repair ability characteristics can be reflected completely, and a whole set of complete reference basis is formulated for scientific assessment of the power distribution network repair ability level.

The invention discloses a power supply area dividing method based on improved gray clustering, including the following steps: S1, establishing an evaluation index system; S2, collecting and preprocessing secondary index data; S3, determining the number of gray classes and the center point thereof; S4, calculating the whitening weight function of each gray class; S5, calculating the integrated weight vector v of the secondary indexes; S6, calculating the clustering coefficient of each secondary index about each gray class; and S7, getting the division result of a power supply area. By improving the most important parts, namely, the index weight calculation method and the whitening weight function design, in gray clustering, the power supply area division result is more accurate. Through reasonable division of the power supply area, optimal construction guidance schemes can be presented for distribution networks after division, in order to improve the utilization rate of assets and promote the overall quality of the power grid.

The invention relates to an otolith measurement and fish population discriminating method based on a polar coordinate, relates to the fish farming and management field and particularly relates to the method for distinguishing fish populations through measuring fish otolith. According to the method, a center point of an otolith picture is taken as a measurement reference point; end pointd, convex concave points or inflection points having prominent features in the otolith form are searched and taken as characteristic points; through establishing polar coordinates, the characteristic points and a core point are connected, radial length between the core point and each characteristic point is measured to acquire a set of characteristic length data; included angles of two adjacent connection lines are sequentially measured by utilizing the measurement software to acquire a set of characteristic angle data; on the basis of the characteristic length data and the characteristic angle data, a discrimination analysis method or a gray cluster analysis method is utilized, a population type of a detected sample is discriminated. Compared with a discriminating method through utilizing only the length, the otolith measurement and fish population discriminating method is advantaged in that the discrimination result is more accurate and reliable, and the fish population discrimination correct rate can be remarkably improved.

The invention provides a reduction method and system for reliability evaluation indexes of a power communication network. A sample matrix of the reliability evaluation indexes of the power communication network is established; a whitenization weight function of all the indexes is established, and a weight coefficient matrix and a synthetic weight coefficient matrix of all the indexes are calculated and acquired; grey clustering is conducted on all sampling points in the synthetic weight coefficient matrix; the clustered synthetic weight coefficient matrix is converted into a decision attribute table, attribute reduction is conducted on the decision attribute table, a minimum attribute subset is acquired, attribute value reduction is conducted on the minimum attribute subset, and a minimum decision rule set is acquired. By the adoption of the method and system, redundant indexes in an index system are removed, the index system can be optimized, analysis efficiency and accuracy are improved, and the power communication network reliability evaluation index system is more reasonable, effective and scientific.

The invention relates to a grey clustering analysis method, comprising the steps of: (1) providing a clustering albetion function: selecting n clustering objects to obtain m clustering indexes and construct an n*m table; (2) inputting the clustering albetion function into a computer and getting into the algorithm of grey clustering analysis and evaluation, comprising: performing equalization nondimensionalization treatment to the clustering albetion; determining the grey type of each clustering index value of each clustering object; calculating the albetion weight function value of each grey type by an estimation method or interpolation method; demarcating a clustering weight grey matrix; and constructing a clustering matrix; and (3) performing grey evaluation according to the result of step (2). The method overcomes the defect of great influence caused by individual factor in single factor evaluation, reflects the influence of the evaluation factors in an all-round way, and has objective evaluation result, convenient and rapid calculation, accurate result, wide applicability, precise evaluation and high sensitivity.

An evaluation system for an intelligent substation comprises an evaluation service platform and at least one terminal center, wherein the evaluation service platform is connected with the terminal center. The terminal center conducts acquisition according to preset data or according to data issued by the evaluation service platform; a four-stage progressive structuralized evaluation index system construction method is adopted for an intelligent substation evaluation index system, wherein the four stages of the construction method refer to a technique layer, an index layer, a criterion layer and a target layer, and the technical advancement degree of the intelligent substation can be reflected. In order to reflect the overall technical development stage of different intelligent substations, the evaluation service platform adopts a grey clustering evaluation method for processing indexes to obtain gray clustering evaluation result values under different grey classes, and a technical development stage qualitative evaluation result and a technical advancement degree quantitative evaluation result of the substation to be evaluated can be reflected according to the gray clustering evaluation result values.

The invention discloses a city water environment carrying capacity grade evaluation algorithm based on the grey clustering analysis. Thus, a more visual evaluation can be directly provided for the grade of the water environment carrying capacity in an area, and a scientific basis is provided for scientifically formulating water environment protection and management decisions in the area. The Delphi expert consultation method is used for determining a water environment bearing force evaluation index system, then an order relation analytical method is used for determining the weights of evaluation indexes, and finally the evaluation indexes of the water environment carrying capacity are divided into four grades by combining with practical conditions and the grey weighing and clustering analysis method is used for determining the grade of the water environment bearing force of the certain area. It is proved by experiments results that the city water environment carrying capacity grade evaluation algorithm based on the grey clustering analysis conforms to the objective practical conditions of the city water environment carrying capacity.

The invention discloses a high-speed railway overhead line system equipment health evaluation method. The method comprises the steps of establishing an overhead line system health evaluation index item set; collecting operation data of the overhead line system, and constructing feature engineering; constructing a training set and a test set by taking the quality evaluation as a label and the secondary index item as a feature dimension, and obtaining a classification model through a classification-random forest algorithm; extracting the weight ratio of each index in the secondary index items inthe classification model; collecting record values of the secondary indexes of the contact network equipment as original data; performing 0-1 standardization on the original data; calculating the weight ratio of each index among the secondary index items; integrating the weight ratios to obtain a final weight ratio of each index in the secondary indexes; constructing a gray class level and a grayclustering function in gray clustering; calculating the coefficient of each gray level to which each contact network object belongs, and taking the level corresponding to the maximum coefficient as the health state of the contact network, thereby achieving the early warning of the health problem of the contact network equipment.

The invention belongs to the technical field of data mining and provides a data supplementation technology which combines the advantages of an expectation maximization algorithm and grey clustering analysis and combines an expectation-maximization method and a clustering analysis method. The data supplementation method ensures the data relevance on the basis of repeated estimation and can be used for completely supplementing data to a missing sequence, thereby improving the calculation efficiency of missing data, shortening the convergence cycle and improving the estimation precision.

The invention relates to a power quality data processing method and device based on intuitionistic fuzzy combination weighting. The method comprises the following steps: acquiring various power quality evaluation index values; dividing comprehensive grades of electric energy quality; carrying out combined weighting based on intuitionistic fuzzy ANP; performing grey clustering processing based on an end point mixing possibility degree function; and determining an electric energy quality grade. According to the method, subjective weights are calculated by applying an intuitionistic fuzzy number to improve a classical ANP scaling method, and the two weights are combined with an entropy weight method in a manner that subjective and objective weighting attributes are consistent. Then, the uncertainty and fuzziness of the grade are processed by adopting a grey clustering method of an endpoint mixing possibility degree function; and finally, a comprehensive clustering coefficient is calculated by combining an index combination weight to obtain an electric energy quality result. Engineers are helped to correctly evaluate the electric energy quality level to a certain extent, theoretical guidance is provided for treating the electric energy quality problem, and the application value and prospect are huge.

The invention discloses a phased array radar resource scheduling scheme evaluation system based on a task profile. The phased array radar resource scheduling scheme evaluation system comprises an index module and an evaluation module, The index module is used for carrying out index hierarchy division on the phased array radar resource scheduling task and setting and calculating evaluation index related parameters of each index hierarchy, and the evaluation index related parameters comprise an index name, an index weight, an index value and an index value normalization value; And the evaluationmodule is used for establishing an evaluation model according to the empirical data and substituting evaluation index related parameters into the evaluation model for calculation to obtain an evaluation result. According to the phased array radar resource scheduling scheme evaluation system, an index module constructed based on a task profile and a gray cluster evaluation module based on a centerpoint mixing possibility function are combined to form the phased array radar resource scheduling scheme evaluation system with completeness and rationality.

The invention discloses a grey clustering evaluation method based on a center point triangle whiten weight function. The method comprises the following steps: setting n decision objects I, m clustering indexes j and s grey classes k, and classifying the decision objects I into the grey classes k according to sample observation values xij of the decision objects I about the clustering indexes j. According to the method, a a grey clustering evaluation model based on the center point triangle whiten weight function is adopted for the performance of suppliers, and the suppliers can be classified and clustered for management; for the model, the improvement of a triangle white function is substantially the replacement of an original pure triangle white function with a hybrid white function (comprising a lower limit measure white function, an upper limit measure white function and a triangle white function), and by such improvement, clustering index valuing range widening problem causing puzzle in the application of the grey clustering evaluation model based on the center point triangle white function for more than twenty years is eliminated.

The invention discloses an all perspective feedback evaluation assessment method based on grey clustering. The method includes the following steps: establishing a college teacher assessment index system, determining the weight of each evaluation index, collecting evaluation data, determining grey clusters and definite weighted functions, calculating definite weighted function values of the evaluation data, calculating a grey fixed weight clustering coefficient, and determining an assessment result. A grey clustering evaluation method is utilized to assess teachers in colleges, so evaluation channels are more comprehensive and fairer, and an assessment result is more scientific and accurate.

The invention discloses a ferrograph for ships and a measurement and analysis method. The ferrograph comprises a ferrogram maker 1, a high pixel camera 2, a USB 3, a virtual instrument 4 and a notebook computer 5 and is characterized in that lubricating oil in an oil sample tank 6 in the ferrogram maker 1 goes through a micro-pump 7, abrasive particles stay on a glass substrate 8, the oil enters into a waste oil tank 9, high definition images of the abrasive particles are taken by the high pixel camera 2, the high definition images are transmitted into the notebook computer 5 by the virtual instrument 4, the images are processed by image pre-processing software in the notebook computer 5, an abrasive particle wear type is determined in a virtual interface of the notebook computer 5 by a grey clustering algorithm, a grey forecasting model (EGM(1,1)) is developed by a grey prediction algorithm and abrasive particle wear tendency is predicted. The ferrogram maker has a simple structure and is suitable for a complex electromagnetic environment of a ship cabin. Through use of the virtual instrument, the ferrograph for ships can be developed and upgraded at any time.

The invention relates to the technical field of LNG (liquefied natural gas) filling station safety evaluation, in particular to a natural gas filling station safety evaluation model and method, and the evaluation model comprises the following steps: S1, building an index system based on a fault tree, S2, determining an index weight based on an analytic hierarchy process, and S3, carrying out safety evaluation on the station based on grey weighting clustering. The evaluation method comprises the following steps: 1, sorting risk indexes influencing the safety of the LNG filling station through design files, field investigation, daily work experience and the like, classifying the risk indexes, and merging the risk indexes into different levels according to classification results; according to the method, the index system is established based on the fault tree, the analytic hierarchy process and the grey clustering method are comprehensively used for calculating the weights and the importance degrees of the influence indexes of the LNG gas station system, the conclusion of safety evaluation of the LNG gas station is more scientific, more accurate and more comprehensive, and the technology is high in practicability.

The invention provides an integrity assessment technology for a three-dimensional physical habitat of an urban river. The technology comprises the following steps of (1) constructing an integrity assessment index system for the three-dimensional physical habitat of the urban river; (2) carrying out field investigation according to an index calculation method in the step (1); (3) constructing an integrity assessment standard system for the three-dimensional physical habitat of the urban river according to a field investigation result in the step (2); and (4) utilizing a modified gray clustering analysis method to carry out integrity assessment on the three-dimensional physical habitat of the urban river. The method adopted in the integrity assessment technology is suitable for integrity assessment of the three-dimensional physical habitat of the urban river, is helpful for recognizing reaches with the poor physical habitat of the urban river and index factors for causing the poor integrity of the physical habitat of the urban river and is applicable to the fields of environmental sciences and ecological restoration.

The invention provides a high-speed train traction system composite fault dynamic risk analysis method based on characteristic quantity. According to the method, a fault chain of a traction motor fault caused by an IGBT open circuit of a high-speed train traction inverter and a small gain coefficient of a speed sensor is analyzed; mathematical modeling is carried out by utilizing a fault mechanismto obtain a curve of the fault rate of the traction motor along with change of tan[sigma], and a composite fault risk chain taking the IGBT junction temperature rise and the motor stator temperatureas characteristic quantities is established. Meanwhile, the risk chain analysis is extended to a specific line scene; risk consequence assessment, the risk consequence is decomposed by using an accident tree analysis method; an evaluation system with transportation operation and maintenance as main consideration factors is established, grey clustering processing is carried out on all index valuesthrough a whitening weight function, risk analysis is carried out on risks caused by composite faults, the risk levels of the risks are obtained respectively, and a theoretical basis is provided for making and implementing emergency measures.

The invention discloses an expressway interweaving area safety evaluation method based on traffic conflicts, and the method comprises the following steps: S1, dividing an expressway interweaving area into a plurality of sections according to the traffic operation characteristics; S2, collecting traffic conflict data of each section, and dividing traffic conflicts into side-click conflicts, rear-end conflicts, general conflicts and serious conflicts; S3, selecting the traffic conflict risk index of each section as a safety evaluation index; and S4, according to the safety evaluation index value of each section obtained in the step S3, determining the safety level of each section of the interlaced area by using a grey clustering analysis method according to an evaluation matrix result. According to the method, the segmented safety evaluation of the expressway interleaving area can be realized, the safety weak part of the expressway interleaving area is effectively identified, the traffic safety of the expressway interleaving area is improved in a targeted manner, and the method has a certain application value in traffic safety management.

The embodiment of the invention discloses a subway station safety assessment method and device, a server and a storage medium. The method comprises the steps of obtaining assessment index data of a subway station; converting the evaluation index data through a triangular fuzzy language variable, and constructing a fuzzy judgment matrix; determining the maximum value of index weight data based on the fuzzy judgment matrix; determining a gray clustering evaluation matrix based on the maximum value of the index weight data; and performing safety evaluation on the subway station based on the grayclustering evaluation matrix. According to the embodiment of the invention, the method and device reduces the impact on the subway safety evaluation from the fuzzy evaluation index, iron out the defect that the grey clustering just considers the grey property of the index, and improve the precision of subway station safety evaluation.

The invention discloses a green characteristic evaluation system for a stone processing process, which comprises an evaluation operation platform and a data terminal center, wherein the evaluation operation platform and the data terminal center are connected; and the data terminal center carries out data acquisition according to a preset index or according to an index issued by the evaluation operation platform. The green characteristic evaluation system for the stone processing process adopts a method of building a three-layer progressive structured evaluation index system, the three layers are an index layer, a criterion layer and a target layer, and thus, the green cleanliness degree of a stone processing enterprise can be characterized and reflected thoroughly. In order to reflect the green characteristic degrees and the grades for different stone processing enterprise production processes, the evaluation operation platform adopts a mode of combining an AHP technology and a grey clustering technology to process the evaluation index and the acquired data, and the evaluation numerical values under different grey classes are extracted. Through the evaluation numerical values, a qualitative green characteristic grade division result and a quantitative green characteristic evaluation data result for the to-be-evaluated stone processing enterprise production process can be characterized, and the green characteristics of the same kind of stone processing enterprises participating in evaluation are ranked. Through an evaluation result output form, problems and gaps among stone processing enterprises can be diagnosed and analyzed, a relevant suggestion can be put forward for implementation and development of a clean production mode of the future stone processing enterprise, and the practicality is strong.

The invention relates to a health assessment method for a hard rock tunneling system, and a sensing technology comprises an intelligent sensing module, a data acquisition module, a remote transmission module and an Internet of Things module. The acquired data are input into the hard rock tunneling system for health assessment, and the method mainly comprises the following steps of 1) selecting the health index information according to a hard rock tunneling intelligent sensing module, carrying out data pre-processing and normalization processing; (2) establishing a health assessment model of the hard rock tunneling system according to the health assessment indexes; (3) selecting the health assessment index data in a period of time; (4) establishing a novel whitening weight function and selecting grey classes according to the index data; and (5) subjectively dividing the health index weights according to an analytic hierarchy process, objectively dividing the index weights according to an entropy weight method, and combining the index weights divided according to the two methods to form the subjective and objective health index weights; and 6) according to the combination of a grey clustering method and a combined weighting method, obtaining a comprehensive grey decision to evaluate the health grade. Compared with the prior art, the method has the advantages of being simple and accurate in calculation and the like.

The embodiment of the invention discloses a lightning arrester quality evaluation method and related equipment. The method comprises the following steps: constructing a lightning arrester quality evaluation system; obtaining a to-be-evaluated lightning arrester sample; on the basis of a grey clustering analysis method, performing first-level fuzzy comprehensive evaluation operation on the lightning arrester sample to be evaluated for all the lightning arrester quality influence indexes to obtain a first-level grey clustering coefficient vector matrix; executing a second-level fuzzy comprehensive evaluation operation according to the first-level grey clustering coefficient vector matrix to obtain a quality membership matrix; and determining the lightning arrester quality grade to which the to-be-evaluated lightning arrester sample belongs in the quality membership matrix based on a maximum membership principle. According to the good lightning arrester quality evaluation method provided by the embodiment of the invention, the quality difference of lightning arresters of different manufacturer models can be selected, a reference is provided for distribution network lightning arrester model selection and optimal configuration, and the fault accident rate and economic loss of a power grid lightning arrester are reduced.