65 results about "Principle of maximum entropy" patented technology

The invention provides a method for early warning of sensitive client electric energy experience quality under the voltage dip disturbance. The method comprises the steps that based on the S conversion rapid algorithm and an increment SVM classifier, voltage dip disturbances of sensitive clients are automatically identified; based on identification results of the voltage dip disturbances, voltage tolerance curves of devices corresponding to multiple types of sensitive clients at different load levels are determined; historical monitoring data of the voltage dip disturbances serve as samples, the samples are converted into sample values of a voltage dip amplitude ponderance index MSI and a lasting time ponderance index DSI, a probability density function of the MSI and the DSI is determined on the basis of the maximum entropy principle, the sensitive device fault probability is evaluated, and the probabilities of the sensitive devices corresponding to the sensitive clients at the voltage dip level are obtained. By the adoption of the method for early warning of sensitive client electric energy experience quality under the voltage dip disturbance, the electric energy quality disturbance condition can be accurately monitored, whether a client load is influenced by the disturbance or not is determined according to the load sensitivity degree of each client, and potential risks of load operation are found.

The invention discloses a linear selecting method in the computing process of hydrological frequency, which includes: firstly, respectively selecting reasonable parameter prior distribution types, initial parameter samples and likelihood functions according to known regional prior information, performing parameter posterior distribution sampling by means of AM-MCMC (adaptive metropolis-markov chain monte carlo) to obtain parameter posterior distribution sampling results corresponding to each line type; analyzing and quantitatively describing probability distribution of the parameter posterior distribution sampling results by means of the POME (principle of maximum entropy) to obtain various parameter posterior distribution formulas in different line types, utilizing the method of approximate summation among application parameters to substitute for linear edge distribution integral process, solving bayes factor Bji of the hydrological line type Mj relative to the line type Mi according to the following formula; and finally selecting and comprehensively analyzing the hydrological line types on the basis of solving the bayes factor Bji. By the linear selecting method, uncertainty of the parameters analyzing and describing are analyzed reasonably, so that analyzed and computed results are improved evidently.

The invention provides an initial-spectrum-independent method for determining neutron energy spectrum in a reactor steady-state neutron field, and discloses a global optimization method for determining neutron energy spectrum by proposing an entropy function for representing neutron energy spectrum distribution and applying a corresponding maximal entropy principle. The initial-spectrum-independent method comprises the following steps of determining a neutron group structure, measuring and recording mononuclear activity of each nuclide, calculating and recording a microscopic cross section of each group of each nuclide, establishing a discrete activation equation set, establishing a probability distributing table of the neutron energy spectrum, and the like. According to the initial-spectrum-independent method disclosed by the invention, the problem that an existing method is dependent on initial spectrum and low in calculating efficiency is solved.

The invention discloses a social network oriented user relationship analysis method, which comprises the following steps: (1) utilizing a web crawler or each social network to open an API (Application Program Interface) platform to obtain original data; (2) extracting the personal attribute of a user, the friend relationship information of the user and the association information of the user, and establishing a personal interest impact factor function, a friend relationship impact factor function and an association-driven impact factor function according to the extracted information; (3) on the basis of a maximum entropy principle, constructing a user relationship analysis module, and carrying out parameter training by the user relationship analysis model to obtain an optimal parameter set; and (4) according to the optimal parameter set and the user relationship analysis module, predicting whether users have a relationship or not. By use of the method, the driving strength of the impact factor can be quantized, and the method also can be used for predicting the development tendency of user relationships and is favorable for finding unknown links and future links in the social network.

The invention provides a portable ultrasonic Doppler fetal heart rate monitor which is safe, is accurate in detection and has multiple functions and a controlling method thereof, belonging to the technical field of medical devices. The fetal heart rate monitor comprises an ultrasonic transmitting and receiving unit, a fetal heart rate signal extraction unit and a fetal heart rate signal processing unit, wherein an output end of the ultrasonic transmitting and receiving unit is connected with an input end of the fetal heart rate signal extraction unit; and an output end of the fetal heart ratesignal extraction unit is connected with an input end of the fetal heart rate signal processing unit. The controlling method of the fetal heart rate monitor comprises the following steps: 1, performing digital filter on a fetal heart rate signal extracted by the fetal heart rate extraction unit by adopting a wavelet packet mixed nose threshold denoising method based on the maximum entropy principle; and 2, adopting a generalized kernel correlated algorithm to performing heart rate detection on the fetal heart rate signal processed in the step 1.

The invention relates to an X-ray image enhancement method based on dual-energy spectrum and wavelet fusion. The method includes the steps of obtaining material specific attenuation changes by changing energy spectrum distribution of an X-ray source in accordance with the principle that attenuation coefficient of a matter is related to energy; selecting the maximum of high frequency coefficient gradients and multiplying the maximum by a weighting factor (the weighting factor is greater than 1) according to the low-contrast feature of X-ray CT (computerized tomography) imaging as a fusion method for increasing high frequency coefficients and highlighting details; calculating the fractal dimension of low frequency coefficients to obtain the complexity, performing image fusion enhancement on dual-energy spectrum two images through objective and subjective weight fusion factors in the maximum entropy principle for overcoming the deficiency that weight selection has no theoretical basis and is too subjective, and making up for deficiencies of a single CT image. The X-ray image enhancement method based on dual-energy spectrum and wavelet fusion is significant.

The invention provides a network health degree testing method and a network health degree testing system. The method comprises the following steps of: acquiring multiple characteristic values of a network; setting preset health degrees for characteristic combinations of the characteristic values respectively; calculating the maximum entropy value of each characteristic combination based on the maximum entropy principle; and taking the preset health degree of the characteristic combination corresponding to the maximum entropy value as the health degree of the network. The sorting operation state of the current network is obtained aiming at the operation characteristic of the current network by acquiring the conventional network states and using the maximum entropy principle and finally revealed in a form of health degree, so that evaluation of the network health degree is realized, subjectivity and inaccuracy of human evaluation are overcome, and the problem of difficult human evaluation under the complex network condition is solved.

The invention discloses a significance testing method for a rolling bearing performance variation process. The method includes: regularly sampling the performance of a rolling bearing during service to obtain an r-th time sequence, equally dividing the r-th time sequence into D subsequences, and setting a d-th subsequence data as Xrd; processing the subsequences by bootstrapping to obtain origin moments of all orders of each subsequence; establishing a probability density function of each subsequence according to the principle of maximum entropy; selecting a reference time sequence; establishing a posterior probability density function; establishing contact ratio; giving a significance level alpha; if the contact ratio is smaller than 1-alpha, determining the d-th subsequence Xrd in the r-th time sequence varies significantly; if not, determining no significant variation occurs. By the method under the condition that probability distribution and trend prior information is absent, variation information of the performance of the rolling bearing during service can be checked timely, failure risk of the bearing is found so that measures can be taken as soon as possible, and serious accidents are avoided.

The invention discloses a load balancing scheduling method based on a maximum entropy principle. The concept of thermodynamic entropy is introduced into load balancing. The internal relation between entropy and load balance is studied. The resource balance distribution is taken as a target. A balanced scheduling problem of the load is equivalent to a maximum entropy problem for solving the system.Meanwhile, a theoretical model and basic properties of the system entropy are given, the entropy is used as a measurement index of the system load balance degree, scheduling and migration are carriedout on the system load according to the entropy change, the system can reach the load balance target faster and better, the response time is shortened, the throughput rate is increased, and thereforethe performance of the cluster system is improved.

The invention discloses a text emotion analysis device and method. The method comprises a data acquisition module, a data preprocessing module and a text class module; the data acquisition module is used for acquiring multiple Chinese texts; the data preprocessing module is used for preprocessing each Chinese text collected by the data acquisition module; the text class module is used for conducting emotion text classification on the Chinese texts processed by data preprocessing; the data acquisition module, the data preprocessing module and the text class module are connected in sequence. According to the text emotion analysis device and method, the device is simple in structure, the efficiency is high, the cost is low, and the analysis result is high in accuracy.

The invention relates to an evaluation method of uncertainty in the entropy measurement for multisensor neural network of poor information, and belongs to measuring and testing field. The method is characterized in following steps:(1)using RBF(Radial Basis Function) to establish a data fusion model of multisensor data conditions under the condition of poor information, acquiring fusion sequence reflecting the measuring process;(2)using the maximum entropy principle to process obtained fused sequence, to obtain the probability density function of the fusion sequence;(3)using obtained probability density function to process true value and regional evaluation, to realize uncertain measurement with poor information. The method allows small sample of measure data, the unknown probability distribution and can make full use of the existing information of measured data to realize effective evaluation of uncertain measurement.

The invention provides a turbine blade probability damage tolerance analysis method. According to the method, the problems of strong nonlinearity of a blade crack propagation life model, large uncertainty and correlation among parameters and the like are solved, and the fatigue crack propagation life distribution of the turbine blade can be predicted only according to the parameter distribution. The method comprises that firstly, a crack propagation life model is logarithmic; Nataf change and orthogonal change are utilized to convert the related random variables into independent normal randomvariables, then two-dimensional dimension reduction integration is utilized to solve the first four-order moment of the crack propagation life, and finally the probability density function (PDF) and the cumulative distribution function (CDF) of the crack propagation life are solved based on the maximum entropy principle (MEP). The method can predict the fatigue crack propagation life distributionof the turbine blade and guide the evaluation of the detection period of the turbine blade.

The invention relates to an attribute interval identification method for intelligent power distribution network efficiency evaluation. A three-layer attribute measuring matrix between intelligent power distribution network basic data and efficiency is built, intelligent power distribution network efficiency attribute intervals are identified automatically on the basis of index data standardization, and attribute measuring interval and comprehensive attribute measurement value of maximum entropy principle calculating performance index, standard index and comprehensive index. The method includes the steps of firstly, reading information required by the evaluation; secondly, building an attribute measuring matrix of comprehensive index, standard index and performance index; thirdly, standardizing index data; fourthly calculating attribute measuring intervals of comprehensive index, standard index and performance index; fifthly, calculating comprehensive attribute measurement of efficiency index, and identifying attribute intervals of intelligent power distribution network efficiency so as to realize intelligent power distribution network efficiency evaluation.

The invention discloses a maximum entropy principle-based shallow sea water flow field acoustic tomography method. According to the method, transmitting-receiving nodes are arranged in a sea area to be tested, so that measured acoustic signals can be obtained; the transmission delay difference of the same path in an opposite direction is obtained; prior flow field information is utilized to construct the relationship of the measured delay difference and flow field distribution, and an inversion mathematical model is constructed; the inversion mathematical model is solved by using the maximum entropy principle; and a convergence solution is obtained by using an iterative method. As indicated by simulation instances, the maximum entropy principle-based shallow sea water flow field acoustic tomography method can be utilized to construct two-dimensional flow field information of the target sea area by using the transmission delay difference and achieve higher least-square accuracy.

The invention relates to the field of computer data mining, in particular to a team competition result prediction method based on a maximum entropy model. The method comprises the following concrete steps of: 1, obtaining feature data of two participating parts of each team in each competition in historical competition seasons; 2, constructing the maximum entropy model according to a maximum entropy principle, and substituting a training data set T into the maximum entropy model to obtain an optimum model; and 3, obtaining an expected substituting optimum model P<*>(Y|X) of each feature datum of the two participating parts in the recent n competitions, and calculating the distribution probability P<*>(Y|X) of the finial competition result of the home field competition participating team. The technical scheme of the invention provides scientific competition result prediction for people concerning the competition.

The invention relates to an optimization method of speech emotion recognition. At present, speech is a tool for communication between people and for thinking and feeling expression; in order to achieve the purpose that a computer can communicate with humans as people, speech emotion recognition gradually becomes a research hot spot in the field of intelligent human-computer interaction; and in China, the research starts relatively late, and a correction rate of the speech emotion recognition is also quite low. The optimization method comprises the following steps: firstly, taking Berlin data set and Mandarin Emotional Speech Database of Chinese Academy of Sciences as speech database of emotion recognition, wherein the speech database includes five emotional speeches, namely happiness, anger, fear, sadness and calm, and recognizing the five emotional speeches so as to select out a test set and a training set; then, implementing characteristic parameter signal extraction on the five emotional speeches, and obtaining SVM kernel parameters from extracted characteristic parameter signals by virtue of a method combining Fisher criteria and the principle of maximum entropy; then, training an SVM by virtue of the SVM kernel parameters; and finally, recognizing speech emotion signals by virtue of kernel parameters which are optimized by the SVM.

The invention discloses a generalized second-category beta distribution based flood frequency analysis method, and belongs to the field of flood prediction in hydrology. The method introducing generalized second-category beta distribution into flood frequency analysis for the first time includes: building a flood probability density predication model, adopting the maximum entropy principle according to distribution function characteristics to realize parameter deduction of the flood probability density predication model, substituting a historical river flood peak sequence into a parameter deduction equation to work out parameters, then substituting the parameters into the flood probability density predication model, and finally utilizing the complete flood probability density predication model to perform probability prediction of massive floods. The flood probability analysis method is extraordinarily suitable for hydrologic frequency analysis, fitting effect is basically superior to that of other traditional distribution in hydrology, and a more effective approach is provided for hydrologic frequency analysis.

The invention discloses a forecast scheduling method for reducing reservoir flood regulation starting water level by considering forecast errors, and belongs to the technical field of flood preventionforecast scheduling. The method comprises the following steps: firstly, performing flood forecasting feasibility analysis on a reservoir control basin, and identifying forecasting error distributionby adopting a maximum entropy principle; secondly, formulating a dispatching rule framework for reducing the flood starting water level of the flood rising section by utilizing a pre-discharge thoughtfor the whole dispatching system; optimizing the forecast scheduling framework to obtain a forecast scheduling scheme optimization point set; thirdly, screening out all optimization point sets meeting the upstream and downstream flood control safety under the condition of the maximum forecast error from the forecast scheduling scheme optimization point set; and finally, comprehensively considering forecast errors and different preferences of a decision maker, and evaluating optimal forecast scheduling scheme points by utilizing a binary comparison method and a fuzzy optimization model to obtain a final forecast scheduling scheme. The method is simple and easy to operate, and the flood control benefit of the reservoir and the elasticity of a downstream protection point under the flood action are improved under the condition of keeping the interest benefit.

The invention discloses a Halphen B distribution-based flood frequency analysis method and system, and belongs to the field of hydrologic analysis and computation. The method comprises the steps of firstly introducing Halphen B distribution in flood frequency analysis; constructing a flood probability density function; realizing parameter estimation of the Halphen B distribution by adopting a maximum entropy principle according to characteristics of a distribution function; and deducing a design flood value in a T-year return period by utilizing the flood probability density function. The invention furthermore realizes the Halphen B distribution-based flood frequency analysis system. The flood frequency analysis method and system is very suitable for hydrologic frequency analysis, and a fitting result of the Halphen B distribution is basically superior to that of other conventional distribution in hydrologic frequency analysis; and a more effective way is provided for the hydrologic frequency analysis.

The invention provides an aero-engine top speed performance digital twinning method based on artificial intelligence. Two sets of deep neural networks with similar structures are adopted, and by meansof an aero-engine numerical simulation performance model established according to the aero-engine principle and the aerodynamic thermodynamic process, the two pain points of low precision and large required data volume caused by neglecting the actual physical process only depending on data of a data driving model are effectively solved. Based on an artificial intelligence deep learning method, anaero-engine brand new performance digital twinning model is constructed, and the training speed and precision of digital twinning are greatly improved by adopting artificial intelligence, a maximum entropy principle acceleration strategy and other key technologies.