56 results about "Evidence reasoning" patented technology

The invention, which belongs to the technical field of the video image, discloses an infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone. The method comprises: step one, an infrared panorama pick-up head is used for carrying out infrared shooting, thereby obtaining a video image signal; step two, an infrared panoramic picture is unfolded by using a fast approximate unfolding method; step three, an indoor grid semantic map in a home environment is constructed; step four, on the basis of an improved hybrid Gaussian model algorithm, modeling of a human body target is carried out, a target block mass is identifier, block mass information is obtained, the target block mass is tracked in real time, and a human body track feature is obtained; step five, a human body moving track is extracted by using a hybrid Gaussian clustering method; step six, the human body track is segmented; and step seven, a convolutional neural network is trained by using the track after human body motion segmentation in the home environment as a training sample, feature extraction is carried out on a human body behavior track in the home environment, and classification is carried out by using evidence reasoning; and if an abnormal situation occurs, alarming is carried out.

The invention discloses a coupling certainty and uncertainty analysis fire risk evaluation method. The method comprises steps that S1, risk factors are identified, and a fire risk evaluation system isestablished; S2, a fuzzy analytic hierarchy method is utilized to determine the index weight of each risk factor; S3, the fuzzy theory is utilized to combine with the confidence to calculate the firerisk of each single factor; S4, the evidence reasoning operator is utilized to synthesize the fire risk of each single factor; S5, an expected utility value is utilized to acquire the clear evaluation result; and S6, the sensitivity analysis method is utilized to verify feasibility of a model.

The invention discloses a trustworthiness dynamic evaluation method facing software evolution. Based on the existing evaluation requirements of the trusted software, and by comprehensively considering the problems that a trustworthiness evaluation index system (TEIS) fails, trustworthiness evidence reasoning can not be developed normally and the like, a self-adaptive reconfiguration unit is designed to implement the TEIS based on associated matrix.. The research and application of the invention provide a convenient and effective solution for the problem of the trustworthiness dynamic evaluation under the background of software evolution, overcome the disadvantages of the conventional models in the aspects of dynamic nature, universality and uncertainty, and ensure the significant value in practical use.

The invention provides a Chinese comment emotion classification method and a Chinese comment emotion classification device based on evidence reasoning rules. The method comprises the following steps of: preprocessing the obtained Chinese comment text; Extracting features of the pre-processed Chinese comment text to obtain a vectorized Chinese comment text; Dividing the Chinese comment text into training data and test data, and training a plurality of preset base classifiers using the Chinese comment text; Predicting the emotion tendency of the test data by using the plurality of trained base classifiers to obtain weights of each base classifier; Based on the weight and reliability of each base classifier, the output probabilities of each base classifier are fused by using evidence reasoning rules to obtain the emotion classification results of Chinese comment texts. Thus, the invention uses evidential reasoning rules to fuse the output results of different base classifiers, and considers the weight values and reliability of different classifiers, so that the final fusion result is more reliable and accurate.

The invention discloses an evidence fusion method for mechanical fault diagnosis of an electric propulsion ship shafting propulsion system. The method comprises the following steps: frequency domain characteristic signals of the positions of a pedestal and a support bracket and the like in different failure modes are obtained; reference value of input characteristics is determined via a K-means algorithm, and similarity distribution of samples is calculated; a casting point statistic table reflecting a relationship between input and failure modes is established, and an input evidence matrix table is obtained via conversion; classification ability and overall uncertainty of an input information source is calculated based on rough set theory and information entropy; reliability and evidenceweight of the input information source are determined; evidence reasoning rules are used to fuse evidence of input sample vector activation, and a failure mode is determined based on fusion results. The method can be used for effectively estimating mechanical failure modes of the shafting propulsion system through vibration signals obtained via a sensor installed on a ship; the method is low in cost and high in precision, and real-time detection and accurate diagnosis of mechanical failure of the electric propulsion ship shafting propulsion system can be realized.

The invention discloses a composite fault diagnosis method based on combination of the artificial immunity and the evidence theory. The composite fault diagnosis method based on the combination of the artificial immunity and the evidence theory comprises the steps that on-line computation of non-dimensional indexes is conducted through vibration monitoring data, and the value range corresponding to the non-dimensional indexes is established during a composite fault; fault diagnosis results obtained through the non-dimensional indexes are converted into credibility distribution by means of the artificial immunity method, diagnosis results of all artificial immunity fault detectors based on the non-dimensional indexes serve as input data through the evidence reasoning algorithm, a final fault diagnosis result serves as output data, a fusion diagnostic model is established, and then composite fault diagnosis is achieved after new input data and new output data are obtained. The composite fault diagnosis method based on the combination of the artificial immunity and the evidence theory fills the blank of a composite fault diagnosis method achieved by means of the advantages of the combination of the artificial immunity and the evidence theory, and real-time and accurate diagnosis of composite faults of rotating machinery is achieved.

The invention discloses a fuzzy logic and evidence reasoning-based transformer insulation stress calculation and evaluation method. The method comprises the following steps of firstly normalizing field data, namely, converting input data for describing a normal value, a limit value and the like of a transformer into dimensionless variables of a [0,1] region; secondly fuzzifying the input data, namely, mapping the input data to a fuzzy membership function, wherein a trapezoidal membership function, a generalized bell-like membership function or a Gaussian membership function can be used; thirdly building a transformer stress evaluation tree model, namely, establishing nodes from the input data of the transformer to an insulation stress of the transformer, and a connection mode of the nodes, and correlating the nodes of upper and lower layers by using a fuzzy correlation matrix; fourthly determining a weight value between the nodes, namely, obtaining a synthesized weight value by using a subjective weight value of an analytic hierarchy process and an objective weight value of eigenvector analysis; and finally obtaining the insulation stress of the transformer, namely, determining a total stress level of the transformer by using a reasoning mechanism and an evidence synthesis rule. The method has the advantages in accuracy, flexibility and capability of processing measurement uncertainty.

The invention discloses a network security situation assessment method based on an evidence reasoning rule. The method comprises the following steps: determining a basic structure of network securitysituation assessment; acquiring and preprocessing network data; determining an assessment rule of the network situation; extracting characteristics of the network data and calculating the trust of a basic attribute; calculating the final trust of a generalized attribute; and calculating the network security situation. According to the network security situation assessment method based on the evidence reasoning rule provided by the invention, a network behavior is divided into a normal behavior and an aggressive behavior; and the security situation of a network is analyzed through the changes of network resources such as traffic, a memory, a CPU and a disk, and the security level of the network is assessed through the above security factors and the evidence reasoning rule. In addition, quantitative analysis is performed on the obtained security situation; and through the change of a network security situation value, the change of the network security situation can be reflected. An experimental result shows that the method is effective and feasible, and has a good situation assessment effect.

The invention provides a method and a device for processing the manufacturing cloud service combination. The method comprises the following steps: screening manufacturing cloud service by utilizing an evidence reasoning method after a manufacturing task is received; selecting manufacturing cloud services fulfilling a designated target from the screened manufacturing cloud service for combining. According to the method, candidate manufacturing cloud services are screened by applying the evidence reasoning method based on screening indexes, the manufacturing cloud service with minimum satisfaction degree not in a preset range can be deleted, and the number of cloud services can be reduced, so that the feasible space of the cloud service combination can be shortened.

The invention discloses an abnormal wear diagnosis method for a marine diesel engine based on union reliability rule reasoning. Firstly, element contents in a working medium lubricating oil of a wearcomponent and a host of the marine diesel engine are used as input, and marine diesel engine abnormal wear component fault conditions are used as output; then, a reliability rule system is establishedand a corresponding attribute weight group is determined according to the correlation between fault features and fault modes; a evidence reasoning algorithm is used to fuse a reliability structure ina reliability rule consequent activated by the input to obtain the reliability value of a fault mode under a specific attribute weight group; and finally, the reliability value is balanced and analyzed to obtain an abnormal wear component diagnosis result. Various elements for marine diesel engine abnormal wear component diagnosis can be handled, a many-to-many complex relationship between the fault features and the fault modes can be characterized, the correlation between the fault features and the fault modes can be characterized, and a complicated condition of concurrence of multiple faultmodes can be diagnosed.

The embodiment of the invention discloses a multi-target optimized dispatch decision evaluation method for a power system. The method solves the technical problem in the prior art that during the current machine group joint optimization scheduling application process, no research is done on the above scientific decision-making method. According to the embodiments of the invention, the method comprises the steps of for a plurality of target functions associated with the optimized dispatch decision of a power system, solving out a Pareto solution set for multi-target optimized dispatch through the multi-target optimization algorithm so as to obtain all candidate solutions; subjecting the attributes of candidate solution sets containing the above candidate solutions to confidence evaluation calculation; according to confidence evaluation, obtaining the overall confidence of the candidate solutions through a multi-hierarchical structure and the evidence reasoning algorithm; and subjecting the total confidence to utility analysis and determining a final scheduling solution.

The invention provides a lake long-time-sequence continuous water area change reconstruction method based on a remote sensing big data platform. Based on a multi-year month-by-month global water bodydistribution data set obtained by a water body extraction scheme integrating an expert system, visual analysis and evidence reasoning in the prior art, through percentage cutting of a water logging frequency image histogram of a researched lake object, an incomplete part in a lake water area inundated area obtained through an existing method is interpolated, and reconstruction of large lake long-time-sequence continuous water area changes is effectively achieved. According to the method, medium-high spatial resolution images (such as land resource satellite Landsat or environmental satellite environmental data) and auxiliary data which are obtained freely can be adopted, the application range of the method is expanded, and an important method support is provided for hydrological analysis and environmental change research of lakes and drainage basins of the lakes.

The invention discloses an evidence reasoning analysis algorithm and entropy weight based air conditioner starting temperature limit value simulation method. The method is implemented by the following steps of: 1) determining an air conditioner starting temperature limit value range and dividing the air conditioner starting temperature limit value range into a finite number of intervals; 2) identifying factors influencing an air conditioner starting temperature limit value; 3) constructing an air conditioner starting temperature limit value interval credibility matrix; 4) calculating weights of the factors influencing the air conditioner starting temperature limit value according to an entropy weight principle; 5) performing air conditioner starting temperature limit value interval comprehensive credibility synthesis with an evidence reasoning analysis algorithm; 6) determining an air conditioner starting temperature limit value interval at a simulation moment and taking a median of the interval as the air conditioner starting temperature limit value at the simulation moment; and 7) repeating the steps 3) to 6) to perform simulation calculation on simulation time periods hour by hour. According to the method, objective factors, subjective factors, quantitative factors and qualitative factors in the air conditioner starting temperature limit value simulation can be processed and incomplete information in the simulation also can be quantitatively processed.

The invention discloses a method for selecting a moving target defense optimal strategy based on DS evidence reasoning. The method comprises the following steps: establishing an attack and defense game model; enabling an attacker to select an attack strategy; enabling a defender to observe the attack strategy, and enabling the attacker and the defender to calculate benefits; refining Bayesian equilibrium solution to obtain an optimal defense strategy; and enabling the defender to correct the type posterior probability of the attacker through DS evidence reasoning. The establishment of the attack and defense game model comprises the steps of defining the attack and defense game model; defining attack and defense revenue quantification. The income calculation step of the attacker and the defender comprises the steps that the defender calculates a posterior probability according to an observed attack strategy; the attacking and defending parties calculate respective revenues; an attack type reasoning decision table is formed according to the characteristic attributes of the previous attack records; and a basic probability assignment of the attack evidence is determined according to the decision table. According to the method, the defect of the prior probability in the existing model is overcome, the profit of the defense strategy is more reasonable, and the defense success rate ofthe early stage of the system is improved.

The invention discloses a railway transportation risk probability calculation method based on a hybrid heuristic rule system. The method comprises the steps that firstly, the environment monitoring quantity, the equipment monitoring quantity and the carrying monitoring quantity of a certain railway transportation interval are collected, and the railway transportation risk probability is output; secondly, reducing the intersection heuristic confidence rule through a reduction rule to obtain a hybrid heuristic rule system comprising the intersection and the union; and finally, an evidence reasoning (ER) algorithm is utilized to fuse the confidence structure in the item after the activated confidence rule is input to obtain the risk probability. According to the invention, various elements ofrailway transportation risk calculation, including environment monitoring quantity, equipment monitoring quantity, carrying monitoring quantity and the like, can be processed; the method comprises intersection rules and union rules, so that the total number of the rules can be reduced, the rule combination explosion problem is avoided, and the calculation efficiency is improved.

The invention relates to a fault diagnosis method for a rotating mechanical ball bearing based on K-means clustering and evidence reasoning. According to the method, a likelihood reliability table andK reference center vectors are obtained by K-means clustering and K reference evidences are obtained by the likelihood reliability table; distances between fault feature data and the K reference center vectors are calculated, the reference evidences are corrected, and K diagnostic evidences are generated; and after on-line acquisition of values of multiple kinds of fault characteristics, activated diagnostic evidences are calculated respectively, the activated diagnostic evidences are fused, a fault decision is made by using the fused evidences, and a fault type corresponding to the online fault characteristic data is obtained. According to the invention, the fusion reasoning of fault diagnosis evidences is performed based on K-means clustering; and with the multi-source diagnostic information, the fault diagnosis accuracy of the rotating mechanical ball bearing can be improved effectively.

The invention discloses a method for fusing measurement data of multi-platform sensors based on D-S reasoning, comprising the following steps: acquiring measurement data of K types of sensors; estimating a system track by adopting an optimal distributed fusion algorithm to obtain accurate measurement data of the K types of sensors; respectively synthesizing feature information of multi-platform same-type sensors for the measurement data of the K types of sensors obtained in the step 2 by utilizing a Dempster-Shafer evidence reasoning method; updating and optimizing the feature information of the multi-platform same-type sensors by adopting evidences of a Jeffrey rule based on a Bayes method, wherein strong condition mass function update evidences and weak condition mass function update evidences are defined in the rule; performing fusion of feature information on multi-platform different-type sensors according to a method which is the same as the step 3 and the step 4; and performing decision output to obtain total fusion data. The method of the invention solves the problem that large calculation amount and low accuracy can be caused if a conventional collaborative conflict resolution method is directly applied to a multi-platform multi-source heterogeneous sensor system.

The invention relates to a ship propeller air entrapment identification method based on evidence reasoning rules and adaptive lifting, and the method comprises the steps: obtaining a three-phase current root-mean-square value and a torque characteristic value from a ship electric propulsion system frequency converter in real time, determining the degree of an air entrapment effect caused by a severe sea condition according to a propeller torque loss coefficient, and setting three levels; firstly, an input feature reference value of each evidence reasoning rule weak learner is given through a K-means clustering method; establishing a fault reliability distribution matrix, converting the input into diagnosis evidence by using the matrix, calculating a reliability factor of the diagnosis evidence, fusing the evidence according to the reliability factor, and estimating the propeller air entrapment effect grade from a fusion result; counting the precision of the current weak learner, and calculating the learning coefficient of the weak learner; and repeating the process to finally obtain a self-adaptive lifting strong classifier composed of a plurality of weak learners, and obtaining anestimated value of the strong classifier for the air entrainment effect level, thereby realizing identification for the propeller air entrainment effect.

The invention relates to a DCS reliability evaluation method based on a dynamic evidence network. The method comprises the following steps of 1) obtaining text data for evaluating the reliability of acomponent, and calculating a probability reliability function of the basic component according to the obtained text data; 2) analyzing a system structure, and establishing the dynamic evidence network; 3) performing evidence reasoning according to the dynamic evidence network, thereby obtaining system reliability; and 4), evaluating the system reliability and the uncertainty. Compared with the prior art, the method has the advantages of simple reasoning method, capability of effectively expressing cognitive uncertainty information, small calculation amount and the like.

The invention discloses an electronic device fault diagnosis method based on grey correlation analysis and improved DS reasoning, and relates to the technical field of fault diagnosis. The invention aims to solve the problem that fault diagnosis and fault prediction capabilities of node test equipment cannot be accurately realized by adopting a simple mode in the prior art. The method comprises: extracting fault characteristics of the tested electronic device to construct a parameter subspace of each fault characteristic; calculating the gray correlation degree of each fault characteristic parameter in the parameter subspace of each fault characteristic and the characteristic parameter in the fault type sample by adopting a gray correlation analysis method; selecting a fault correspondingto the characteristic parameter with the maximum correlation degree from the plurality of grey correlation degrees as a preliminary fault type of the tested electronic device; and performing probability assignment on each gray correlation degree to obtain a basic probability assignment of each fault type, and fusing the basic probability assignments by adopting an improved DS evidence reasoning fusion method to obtain a final fault type of the tested electronic device. The method is used for diagnosing electronic device faults.

The invention discloses a bearing fault diagnosis method for an oil transfer pump, which is characterized by comprising the steps of (1) performing data acquisition on bearing vibration signals by adopting acceleration sensors, carrying out EMD (Empirical Mode Decomposition) and intrinsic mode function (IMF) envelope spectrum analysis, and obtaining sample data characteristic physical quantities;(2) processing the sample data characteristic physical quantities based on a bearing fault characteristic matrix to obtain BPA (Basic Probability Assignment) values to serve as BPA evidences; and (3)improving D-S evidence reasoning rules to fuse the BPA evidences so as to obtain a bearing fault diagnosis result. The bearing fault diagnosis method provided by the invention for the oil transfer pump can process a circumstance that partial data loses efficacy or is abnormal because of the existence of damaged sensors in the multiple bearing acceleration sensors.