33results about How to "Quantitative uncertainty" patented technology

The invention relates to the field of pattern recognition, and discloses an unbalanced data classification method based on intuitive fuzzy integration and a system based on the method. The method comprises the following steps of: a) cleaning original data, and classifying original point-of-sale (POS) class samples according to intra-class positions to generate POS class artificial samples; b) training a base classifier by using different sample sets of inter-class approximate balance; c) converting the classification output equal utility of the base classifier into an intuitive fuzzy matrix; and d) integrating samples to be classified into the membership and the non-membership of the POS class and the negative (NEG) class by combining the weight of the base classifier, and making a classification decision. The invention has the advantages that: over learning is avoided by integrating over sampling and under sampling; the training samples of the base classifier are different, so that the difference of the base classifier is ensured; the base classifier is not specifically limited, so the method has good expandability; the intuitive fuzzy reasoning method quantitatively describes the uncertainty in classification so as to improve the performance of integrated learning; therefore, the system based on the method can better support the medical diagnosis decision and the like.

The invention discloses a wind power climbing event probability prediction method and system based on a Bayesian network, and the method comprises the steps: mining the dependency relationship betweena wind power climbing event and related meteorological influence factors such as wind speed, wind direction, temperature, air pressure, humidity, and the like, and building a Bayesian network topological structure with the highest fitting degree with sample data; quantitatively describing a conditional dependency relationship between the climbing event and each meteorological factor, estimating the value of each conditional probability in a conditional probability table at each node of the Bayesian network, and forming a Bayesian network model for predicting the wind power climbing event together with a Bayesian network topological structure; deducing the probability of occurrence of each state of the climbing event according to the numerical weather forecast information of the mastered prediction time; the value of the corresponding conditional probability at each node is adaptively adjusted, so that the inferred conditional probability result of each state of the climbing event is optimized, and the compromise between the reliability and the sensitivity of the prediction result is realized.

The invention relates to an algorithm for correcting structure model parameters based on a frequency response function, which comprises the following steps of: acquiring time history data and time history response data, and introducing a multivariate circle symmetry proportion distribution theorem to derive and obtain a probability density function and a covariance matrix of the actually measuredfrequency response function; Introducing a prediction error and a to-be-corrected parameter to obtain a covariance matrix containing the to-be-corrected parameter; Obtaining a probability density function of a frequency response function under the action of single-point excitation according to the determinant and the inverse theorem of the matrix; Obtaining a maximum likelihood function expressedin a form of a maximum likelihood function and a logarithm maximum likelihood function according to a maximum likelihood principle; Obtaining a posterior probability density function of the random variable according to the Bayesian theorem; And expressing the posterior probability density function as a logarithm likelihood function form, so that an objective function is obtained. The uncertainty of the correction parameters is quantized, the calculation precision of the correction parameters is improved, and the correction of the structure finite element model is realized.

The invention belongs to the technical field of runoff prediction, and discloses a runoff probability prediction method and system based on deep learning, and the method comprises the steps: employinga maximum information coefficient to analyze the linear and nonlinear correlation between variables, so as to screen a runoff correlation factor; building an extreme gradient boosting tree model on the basis of correlation analysis, and inputting runoff correlation factors into a trained XGB model to complete runoff point prediction; inputting a point prediction result obtained by the XGB model into a GPR model, and performing secondary prediction to obtain a runoff probability prediction result; selecting confidence and acquiring a runoff interval prediction result under the corresponding confidence through Gaussian distribution; and optimizing hyper-parameters in the XGB model and the GPR model by adopting a Bayesian optimization algorithm. A high-precision runoff point prediction result, an appropriate runoff prediction interval and reliable runoff probability prediction distribution can be obtained, and the prediction method plays a crucial role in utilization of water resourcesand reservoir scheduling.

The invention relates to a distributed resource transaction method of a microgrid community under a multi-agent framework. The method includes three iterative electricity price clearing layers. The first layer is used for operating optimization calculation in the microgrid. The second layer is used for operating scheduling optimization calculation in an aggregator. The third layer is used for transaction calculation between the aggregators. Each of the three iterative electricity price clearing layers performs real-time electricity price settlement, and achieves, from local to overall, benefit maximization of the microgrid, the aggregators, and the microgrid community respectively. In each phase of the iteration, distributed resource owners and the aggregators can calculate profits by the method in the present invention, thereby making the market clearing process clearer. The mode optimizes the configuration of distributed resources within the microgrid community while ensuring the benefits of all market participants, thereby ensuring the safety of each microgrid, ensuring the reasonable income for the distributed resource owners, and enhancing the overall balance capacity of renewable energy by price excitation.

The invention provides a high-speed rail track response prediction method based on sparse Bayesian width learning. The method comprises the steps: carrying out the linear and nonlinear feature extraction of an input temperature field variable, carrying out the maximum posteriori estimation of a hidden layer neuron node output layer weight, predicting a structure response output result, and evaluating the structure state of the track preliminarily. According to the method, the sparse Bayesian width learning method is adopted to carry out correlation mining on the data of the high-speed rail monitoring system, and the over-fitting problem of regression prediction can be effectively avoided through sparse solution of the weight w reflecting the relation between data variables. The method has the advantages of high prediction precision, high calculation speed and loose equipment hardware requirements, so that the mining of the correlation between the temperature load and the structural strain implied in a large amount of monitoring data can be realized, and meanwhile, the evolution of a monitoring data model is found in time to serve as a basis for judging the abnormal service state of the track structure.

A method is disclosed for generating pixel risk maps for diagnostic image reconstruction. The method produces uncertainty/variance maps by feeding into a trained encoder a short-scan image acquired from a medical imaging scan to generate latent code statistics including the mean μ_y and variance σ_y; selecting random samples based on the latent code statistics, z˜N(μ_y,σ_y²); feeding the random samples into a trained decoder to generate a pool of reconstructed images; and calculating, for each pixel of the pool of reconstructed images, the pixel mean and variance statistics across the pool of reconstructed images. The risk of each pixel may be calculated using the Stein's Unbiased Risk Estimator on the input density compensated data, that involves calculating the end-to-end divergence of the deep neural network.

The invention discloses a photoelectric probability density prediction method based on B-spline quantile regression, and the method comprises the steps: 1, collecting photoelectric data, carrying outthe normalization of the photoelectric data, and dividing historical photoelectric data into a training set and a test set; 2, constructing a B-spline quantile model, and calculating parameters of theB-spline quantile regression model by utilizing the training set data; and 3, substituting the test set data into the B spline quantile model to obtain predicted values under different quantiles, andrealizing photoelectric probability density prediction by applying kernel density estimation. The prediction precision of photovoltaic power generation can be improved, and the uncertainty of a prediction result is comprehensively measured, so that a reliable basis is provided for safely and stably connecting the photovoltaic power generation into a power grid.

The invention discloses a power system comprehensive decision-making method based on double-layer fuzzy optimization, belonging to the field of power system management. The invention provides a hierarchical decision-making method for electric power production and environmental management in electric power system planning, considers the fuzzy uncertainty of economic parameters in the system and uses a double-layer fuzzy programming method to solve the problem. In the process of solving, the decision makers at different levels control different decision variables and optimize their own objectives. Decision-makers at the upper level make decisions first, and the lower level reacts according to the decision-making results at the upper level. Therefore, the optimization method disclosed by theinvention obtains the optimal power system planning scheme by balancing the benefits of the two layers, and the optimization result disclosed by the invention can provide technical support for studying the regional power system planning, the comprehensive utilization of resources and the environmental protection.

The invention relates to an oil reservoir geologic modeling static parameter distribution prediction method based on a neighbor neural network, and the method comprises the following steps: S100, selecting an oil reservoir geologic prediction range, and calling the known space coordinates and corresponding static parameter values of all wells in the oil reservoir geologic prediction range; s200, finding an adjacent well of each well by adopting a neighbor algorithm; s300, establishing a neural network model and training the neural network model; and S400, predicting static parameter distribution of unknown space points in the selected oil reservoir geological prediction range by utilizing the optimal neural network model obtained by training. The method makes full use of the excellent ability of a neural network to approach a complex nonlinear function, can deeply excavate the nonlinear distribution relationship of static parameters in space, accords with the complex characteristics ofoil reservoir geology, can improve the precision of spatial interpolation, can also quantify the uncertainty of spatial interpolation through multiple random implementation, and the static parameterdistribution prediction precision is improved.

The invention relates to a sensor arrangement method for reducing structural vibration mode matrix recognition uncertainty, which is used for civil engineering structure health monitoring and is characterized by comprising the following steps: arranging a posterior probability density function of a vibration mode matrix and a sensor based on information entropy; the posterior probability density function of the vibration mode matrix is used for describing the uncertainty of vibration mode matrix recognition; according to the sensor arrangement based on the information entropy, the uncertainty of vibration mode matrix parameter recognition is quantified according to the information entropy criterion, and arrangement of the sensors is guided through the information entropy criterion by means of a sequence algorithm.

The invention aims to solve the problems that an existing electricity market electricity price prediction method consumes too long time, more samples cannot be trained to improve the prediction precision and mainly has limitations that research on point prediction, electricity price interval prediction and probability density estimation is less, and more effective information cannot be provided for market participants to make an optimal market strategy. An electricity price prediction method based on empirical mode decomposition and minimum gated memory network quantile regression is established, the electricity price sequence is decomposed into a plurality of modal components by adopting empirical mode decomposition; QR and MGM are combined to form a hybrid model QR-MGM to predict each modal component under different quantiles, and KDE is adopted for a reconstructed prediction result to obtain a probability density function of the electricity price, so that the uncertainty of prediction can be quantified, and the probability density function of the electricity price can also be obtained.

The invention relates to a dietary heavy metal chronic exposure assessment method based on beta binomial normal distribution, and the method mainly comprises the steps: calculating the intake of pollutants of an investigated crowd in the number of days of investigation through the consumption data obtained through investigation and data obtained through literature retrieval or pollutant monitoring, and then building a linear random effect model, the variation of dietary survey is decomposed into intra-human variation and inter-human variation, the positive intake is calculated after the intra-human variation is removed, finally, the dietary frequency of the crowd is obtained through computer simulation and sampling, the dietary frequency is multiplied by the usual intake calculated in the first part, and the long-term intake level of the crowd intake pollutants is obtained. According to the method, a probability statistical method is adopted, the method can be applied to semi-probability or total probability calculation, and uncertainty and variability in the calculation process can be quantitatively counted.

The invention discloses a TVF-EMD-MCQRNN load probability prediction method based on fuzzy C-means clustering. The method comprises the following steps: 1, preprocessing power loads and influence factors thereof, and grouping the preprocessed power loads and influence factors thereof according to a set time interval; 2, dividing the grouped data sets into training sets and test sets, and respectively clustering each group of training sets and test sets by using a fuzzy C-means clustering method; 3, training and testing the TVF-EMD-MCQRNN model by utilizing the various types of training set andtest set sample data to obtain a series of conditional quantiles of the various types of sample data under different quantiles; 4, summing the conditional quantiles of each type of sample data underthe same quantile to obtain each type of prediction result, thereby realizing probability density prediction. According to the method, the accuracy of load prediction can be improved, and therefore more comprehensive and more effective load information is provided for operation planning of a power system.

A sand-mud interbed type reservoir developmental zone earthquake prediction method based on big data analysis comprises the following steps: 1, building a mass sand-mud interbed type geologic model based on the deposition characteristics of a drilled reservoir, and obtaining a transition probability matrix; 2, constructing a 0-degree incident PP wave reflection seismic record of the model in a forward modeling algorithm and a 90-degree phase Ricker wavelet simulation model; 3, screening a model trace data set which is highly matched with the seismic record waveform of the actual stratum through a big data analysis technology, and carrying out normalization processing, and matching a record closest to the actual seismic waveform in the big data model according to a formula; and 4, obtaining quantitative prediction and characterization of the reservoir dominant developmental zone earthquake. According to the invention, seismic forward modeling is realized; moreover, the probability density distribution of the sand-to-ground ratio corresponding to the corresponding geologic model and the maximum single-layer sand body thickness is obtained through statistics, the reservoir development degree is indicated, and meanwhile, the uncertainty of interpretation is quantified.

The invention relates to a sensor arrangement method for reducing structural damping ratio recognition uncertainty, which is used for civil engineering structure health monitoring and is characterized by comprising the steps that a posterior probability density function of a damping ratio and a sensor based on information entropy are arranged; the posterior probability density function of the damping ratio is used for describing the uncertainty of damping ratio identification; according to the sensor arrangement based on the information entropy, the uncertainty of damping ratio parameter identification is quantified according to the information entropy criterion, and arrangement of the sensors is guided through the information entropy criterion by means of a sequence algorithm.

The invention discloses a screening method and device for green development schemes of shale gas. The method comprises the steps of S110, building a multi-layer target optimization model according to the actual condition in a shale gas exploitation process, wherein the multi-layer target optimization model comprises an upper-layer production target model, a middle-layer environment target model and a lower-layer economic target model; S120, solving the upper-layer production target model, the middle-layer environment target model and the lower-layer economic target model to obtain a result of the upper-layer production target model, a result of the middle-layer environment target model and a result of the lower-layer economic target model; S130, judging whether the result of the upper-layer production target model, the result of the middle-layer environment target model and the result of the lower-layer economic target model are equal or not, if yes, ending the process, and if not, entering the step S140; and S140, performing comprehensive solution on the upper-layer production target model, the middle-layer environment target model and the lower-layer economic target model by using an interactive solution algorithm of fuzzy satisfaction to obtain an optimal solution of the multi-layer target optimization model.