32 results about "Uncertainty calculation" patented technology

The invention provides a wave-rider aircraft aerodynamic data uncertainty analysis method and device, and the method comprises the steps: obtaining the numerical calculation aerodynamic data of a wave-rider aircraft; factors influencing the uncertainty of the numerical calculation pneumatic data are determined; carrying out uncertainty analysis on the basis of pneumatic data, calculating model uncertainty by adopting a range method, designing an orthogonal test according to factors, calculating numerical value and input parameter uncertainty, and carrying out significance analysis on factor influence and interaction according to the orthogonal test; and calculating the total uncertainty according to the model uncertainty, the numerical value and the input parameter uncertainty, and verifying the reliability of the total uncertainty by adopting new sample data. According to the method, the uncertainty of the pneumatic data is comprehensively and effectively obtained, meanwhile, the influence factors of the uncertainty of the numerical calculation pneumatic data and the significance of the interaction are visually obtained, and support is provided for the reliability of the numerical calculation pneumatic data.

This invention relates to a method for constructing a machine learning potential energy model based on hierarchical active learning, comprising the following steps: constructing a database of background solvent molecules and lithium salt molecules; initializing a machine learning potential energy model committee; outer active learning automatically exploring the chemical species composition space of the electrolyte, generating electrolyte formulations and initial structures; determining whether a structure needs to be included in the labeling range based on uncertainty calculations; automatically constructing a liquid phase environment based on classical force field simulation methods; performing high-precision simulations based on first-principles calculations and labeling the energy and force values ​​of the structures; inner active learning driving the machine learning potential energy model to explore the configuration space and label the energy and force values ​​of structures with high uncertainty; obtaining a first-principles structure database of the real liquid phase environment through several iterations; and training to obtain the final high-performance machine learning potential energy model. This invention systematically and comprehensively samples the chemical species composition space of the electrolyte and the molecular simulation configuration space based on hierarchical active learning, thereby ensuring the predictive performance and simulation accuracy of the machine learning potential model on unknown and complex electrolyte systems.

The invention relates to the technical field of automobile electronic safety, in particular to a vehicle-mounted abnormity comprehensive detection method and system based on evidence deep learning. The evidence deep learning model in the invention comprises four layers of core structures. The feature input layer receives information entropy of a vehicle-mounted CAN message sequence; model parameters are built in the evidence reasoning layer, and input features are mapped into basic probability distribution corresponding to a plurality of categories; the uncertainty calculation layer quantifies uncertainty measurement in the classification process; and the decision output layer fuses the basic probability distribution and the uncertainty measurement result and outputs a classification result. In the training stage, a loss function formed by a multilayer perceptron evidence function and a product of a preset coefficient and uncertainty measurement is adopted, and in the verification stage, on the basis of the loss function in the training stage, an optimal annealing factor is added for weight adjustment. The optimal annealing factor is determined through correct classification on a maximized verification set and the proportion of samples with the uncertainty lower than an annealing threshold value, and the classification precision is effectively improved.

The invention relates to the technical field of computer security, in particular to a dynamically updated Android malicious software continuous learning detection method, which comprises the following steps of: acquiring and sequencing historical Android application samples according to time to construct a training set, and training a hierarchical comparison classifier; new applications are collected regularly to form a to-be-tested batch, and the current classifier is used for prediction; for each sample in the to-be-detected batch, executing pseudo loss uncertainty calculation to obtain an uncertainty score of the sample; according to the score, selecting a predetermined number of most uncertain samples for labeling, and obtaining a real label; adding the new labeled sample into the training set, and performing incremental training by adopting a hot start mode based on the weight of the current classifier to obtain an updated classifier; and circularly executing, and detecting the new application in the next period by using the updated classifier. According to the method, the labeling cost can be remarkably reduced, the concept drift can be effectively coped, and efficient, stable and continuous malicious software detection is realized.

The application discloses a kind of based on multi-modal data fusion fermentation environment control method, comprising: collection multi-modal fermentation data and executes pre-processing, forms multivariate time series dataset;Improved MTAD-GAT model is input, calculates prediction error reconstruction error and consistency score, generates dynamic reliable weight;Coupling relationship matrix is constructed across mode, coupling residual is calculated and iterative correction data, reconstructs state vector and uncertainty;Coupling stability index is calculated, and state anchoring is executed, obtains stable reference state parameter;Based on state vector and reference state, construct metabolic feasible region, form control constraint interval;Self-scaling robust controller is constructed, dynamic scaling constraint and rolling optimization calculation, output fermentation control instruction.The application realizes stable identification to fermentation process state and adaptive control to fermentation environment parameter by multi-modal data fusion, coupling relationship analysis and self-scaling robust predictive control.

The invention relates to the technical field of performance evaluation and efficiency calculation of a photovoltaic power generation system, in particular to a method and system for calculating the uncertainty of the operation efficiency of a photovoltaic system, and the method comprises the steps: obtaining real-time operation parameters, such as system output power, assembly temperature, irradiance and the like; constructing a nominal power calculation function and calculating the sum of nominal power generation powers of the components at corresponding moments; obtaining an instantaneous performance ratio, and respectively calculating four types of uncertainty components, namely system output power, a temperature correction coefficient, component nominal power and system irradiance; solving the uncertainty of the synthesis standard and the effective degree of freedom of the synthesis standard through a variance synthesis formula, obtaining the expansion uncertainty and the relative expansion uncertainty in combination with the target confidence, and finally achieving the real-time quantitative evaluation of the health state of the distributed photovoltaic system. The method effectively reduces calculation errors under complex working conditions, solves the problems of evaluation lagging and insufficient accuracy of a traditional method, improves data reliability, and adapts to real-time evaluation requirements.

The invention relates to the field of electric energy metering, and particularly provides an uncertainty evaluation method, device and equipment for a measurement result of electric power measurement equipment. The method comprises the steps of collecting a voltage data flow in a current sampling period, and calculating a current power grid parameter based on the voltage data flow; wherein the power grid parameters comprise a voltage total harmonic distortion rate, a voltage fluctuation rate, a frequency offset and a voltage three-phase unbalance degree; determining a current power grid state grade according to the current power grid parameters; when the current power grid state is in a stable state, selecting a target uncertainty calculation model and a target inclusion factor calculation model corresponding to the power grid state grade, and calculating the current synthesis standard uncertainty and the current inclusion factor; based on the current synthetic standard uncertainty and the current inclusion factor, a current extended uncertainty is calculated to evaluate a measurement result of the power measurement device according to the extended uncertainty. According to the technical scheme provided by the invention, the accuracy of uncertainty evaluation can be improved.

The invention relates to a machining uncertainty calculation method, in particular to a machining uncertainty calculation method of a technological process composed of multiple machining procedures. The method is characterized in that the process processing uncertainty is the synthesis of the regeneration processing uncertainty and the genetic processing uncertainty. The machining uncertainty of each process is a machining uncertainty vector composed of one or more machining uncertainty components. The genetic processing uncertainty is a Hadamard product of the processing uncertainty of the previous process and a genetic coefficient. The regeneration processing uncertainty is the combination of the uncertainty of all error sources and the uncertainty of the Hadamard product of the corresponding sensitivity. The uncertainty of the error source is the uncertainty synthesis of all the uncertainty components of the error source and the Hadamard product of the corresponding sensitivity. The method has the advantages that before the process test is implemented, whether the workpiece precision level which can be achieved by the given process meets the design drawing requirement or not is quantitatively calculated and predicted, the process reasonability and effectiveness are evaluated, the process test risk and cost are reduced, and the first workpiece trial-manufacturing success rate is increased.

The invention relates to the technical field of surveying and mapping, and discloses an image-control-free oblique photography surveying and mapping method and system fusing air-ground multi-source data, and the key points of the technical scheme are multi-source space-time alignment and terrain adaptive optimal flight height calculation based on hardware pulse per second; carrying out downsampling acceleration and three-dimensional uncertainty calculation considering the space-ground scale; 2D-3D rigorous physical projection of high-order distortion and drift compensation is integrated; the three-dimensional uncertainty based on the Jacobian matrix is strictly propagated to a two-dimensional image domain; carrying out robust beam method combined adjustment by fusing terrain self-adaption and mahalanobis distance; marginalized accelerated solution and live-action three-dimensional model generation are carried out; a space-time synchronization and space uncertainty cross-modal propagation technology under adaptive flight height constraint is adopted, and a strict physical projection model and terrain adaptive robust joint adjustment algorithm is combined, so that multi-sensor hardware drift is accurately eliminated, and cross-modal mismatching outer points are effectively eliminated; therefore, image-control-free high-precision live-action three-dimensional surveying and mapping in a complex scene can be realized.

The invention discloses a Doppler differential interferometer phase uncertainty calculation method based on spectral domain noise analysis, and solves the problems that the quantitative evaluation of the phase uncertainty of a Doppler differential interferometer needs to pass through the signal-to-noise ratio and the modulation degree of interference fringes, and a single-frame interference pattern in actual measurement cannot be used for calculating the phase uncertainty of the Doppler differential interferometer. The signal-to-noise ratio and the modulation degree of an interferogram are difficult to directly obtain, so that significant deviation exists in phase uncertainty evaluation; according to the method, the phase uncertainty model is constructed by combining the frequency domain noise characteristics of the interference fringes with the interference intensity of the complex interference pattern, and the phase uncertainty is directly quantized; and representing the noise characteristics at the characteristic spectrum by using the statistical characteristics of the noise frequency band outside the characteristic spectrum. The noise variance of the real part and the imaginary part of the complex interferogram is deduced in combination with the Parseval theorem and the error propagation theory, and the phase uncertainty is finally obtained in combination with a phase uncertainty model. According to the method, the defects of traditional indirect evaluation are avoided, and the phase uncertainty precision can be effectively improved.

The invention discloses a disease risk prediction result confidence evaluation and credibility division method, and belongs to the technical field of disease risk prediction.The disease risk prediction result confidence evaluation and credibility division method comprises the following specific steps that firstly, medical feature data of a target patient is obtained, inputting a pre-trained disease risk prediction model to obtain a disease risk probability value of the patient; 2, calculating a model ambiguity confidence factor MCF based on the model integration uncertainty, generating N sub-models through N-fold cross validation, and obtaining a prediction probability value of each sub-model for a target patient; according to the method, a model ambiguity confidence factor (MCF) and a statistical confidence factor (SCF) of a machine learning model are combined, and the comprehensive confidence of a prediction result of each individual is calculated. The fusion method is more comprehensive than a mode of purely depending on a model output probability or a single uncertainty index, and can reflect the reliability of a single prediction result more accurately.

The invention provides a calculation method, device and equipment for pixel-by-pixel uncertainty of a remote sensing product, and relates to the technical field of remote sensing, and the calculation method comprises the steps: determining the respective uncertainty of a plurality of remote sensing products related to each target pixel for a plurality of target pixels in a target remote sensing image; generating a plurality of two-tuples related to the target remote sensing product based on the value of the uncertainty of each remote sensing product and respective parameters of the target remote sensing product related to each target pixel; performing regression analysis based on the plurality of two-tuples related to the target remote sensing product to obtain an uncertainty calculation model related to the target remote sensing product; and performing uncertainty calculation on a plurality of pixels included in the target remote sensing image by using an uncertainty calculation model related to the target remote sensing product.

The invention provides a high-entropy alloy structure performance collaborative prediction method adopting an active learning strategy, and belongs to the technical field of machine learning application. The method solves the problems that the prior art depends on a fixed data set, cooperative and accurate prediction of the structure performance of the high-entropy alloy cannot be achieved under the conditions of sample scarcity and uneven distribution, and the generalization ability is weak. The technical scheme comprises the following steps: constructing and preprocessing a data set; constructing a feature set based on the properties of materials, and determining a key feature subset through multi-stage screening; training an initial collaborative prediction model; an active learning strategy is adopted, and high-value samples are screened through uncertainty calculation and clustering analysis for experiments; and an experimental result is fed back to the model for iterative optimization. According to the method, the prediction precision and generalization ability of a complex component space are remarkably improved, the model interpretability is enhanced, continuous autonomous optimization under the small sample condition is achieved, the method is suitable for various high-entropy alloy systems, and the research and development efficiency of materials can be greatly improved.

The present application relates to the technical field of intelligent monitoring of electric power metering equipment, in particular to a mutual inductor full life cycle uncertainty tracking method and system based on physical information neural network. The method realizes real-time tracking of uncertainty through data preprocessing, data noise reduction processing, physical equation definition, physical information neural network construction and uncertainty calculation. The present application is economical and practical. A system framework based on physical information neural network is proposed. Through the combination of the full life cycle monitoring of the data analysis module, the calibration efficiency and result reliability are improved. Not only can the limitations of traditional technology be solved, but also the uncertainty tracking can be optimized through physical constraints, providing more accurate full life cycle management, significantly reducing maintenance costs and improving system stability.

It is evaluated how reliable a determination result by a determination model is. A determination evaluation device that evaluates a determination result for input data based on a determination model includes an uncertainty calculation unit that calculates uncertainty of the determination result, and an analysis evaluation unit that evaluates the determination result on the basis of the uncertainty.

This invention discloses a PM based on the CMB model. 2.5 An online source apportionment method and device can automatically identify pollution sources without requiring a large amount of receptor data. The method includes the following steps: acquiring historical observation data of particulate matter and its chemical components to be monitored at the monitoring site; selecting historical component concentration data for constructing a PMF model based on the historical observation data, using mass spectrometry information m / z 44 as the SOA source identifier component, and calculating the uncertainty corresponding to each concentration data based on a predetermined uncertainty calculation method; inputting the data into the PMF model and performing calculations to obtain the component spectrum data of each pollution source at the monitoring site and its corresponding uncertainty; acquiring real-time observation data of the chemical components of particulate matter at the monitoring site; filtering the real-time observation data and performing calculations based on a predetermined uncertainty calculation method to obtain real-time component concentration data and its corresponding uncertainty; inputting the data into the CMB model and performing calculations to obtain the impact of each pollution source on PM2.5. 2.5 Contribution results.

The application provides a balance calibration uncertainty calculation method, considers the contribution degrees of a balance calibration system, systematic errors and random errors to the uncertainty, and comprehensively calculates the uncertainty of the balance static calibration according to the influences of the three parts on the uncertainty. The application intuitively gives the contribution of each factor in the balance calibration system to the calibration uncertainty, and facilitates researchers to measure the quality of each part of the calibration system. Meanwhile, the overall structure of the calibration system and the calibration method mentioned in the application is the same as the mainstream calibration system and calibration method, so that the calculation method can be partially referenced by other calibration systems, and has certain universality.

The invention provides a power transmission line local wind field deviation calibration method, device and system. The method comprises the following steps: acquiring multi-source wind field data; constructing an anisotropic covariance model reflecting wind field directivity correlation and terrain non-stationary characteristics; generating background wind fields on the buffer zone grids on the two sides of the power transmission line by using wide-area observation data; correcting a background wind field by point location observation data, performing data fusion and deviation calibration based on an anisotropic covariance model, and outputting a calibrated wind field value of each grid point in the buffer zone grid and the uncertainty of the calibrated wind field value; according to the calibrated wind field value and the uncertainty thereof, calculating an overproof probability that the wind field along the power transmission line exceeds a preset equipment toughness threshold; the risk level map is generated based on the over-standard probability, the problem that the precision of power transmission line local wind field calibration is insufficient under the complex terrain is effectively solved, and the accuracy of buffer area wind field space distribution is remarkably improved through multi-source data fusion and anisotropic modeling.

This invention discloses a method, device, storage medium, and program product for monitoring the concentration of galvanized baths. The method includes: acquiring multispectral data collected by an optical fiber probe; calculating a quality assessment index vector and extracting feature vectors; calculating the reliability weights of each spectrum based on current measurement conditions using an adaptive attention module; weighting and fusing the features of each spectrum according to the reliability weights; inputting the fused features into a neural network and adding random perturbations to perform multiple forward propagations to obtain multiple sets of predicted values; statistically calculating the mean as the final predicted value and the standard deviation as the uncertainty; calculating the deviation between the predicted temperature and the measured probe temperature and adjusting the reliability score accordingly; and assigning a reliability level based on the uncertainty and reliability score. This invention avoids contamination by inferior data through adaptive weight adjustment and provides a reliability assessment by combining uncertainty quantification and temperature cross-validation, thereby improving monitoring reliability and accuracy.

The application discloses an intrusion detection method based on information entropy theory and a convolutional neural network, which comprises the following steps: firstly, converting character type data into numerical type data, data standardization and data normalization operation are performed on a data set; then, the data set is put into a convolutional neural network for dimension reduction and classification, and information entropy uncertainty calculation is combined to perform delayed relearning classification decision on part of the data, and a random forest method is selected as the delayed decision method; when an intrusion behavior occurs, a trained model can be used to distinguish normal data and attack data. The application utilizes the characteristics that the convolutional neural network has better feature extraction capability and classification learning effect, and combines information entropy theory to evaluate the classified data, and the evaluation result is used as the basis for secondary learning classification decision, so that the method can avoid the risk of misclassification caused by insufficient information extraction as much as possible, and the performance of intrusion detection is improved.

The application discloses a kind of based on graph convolution model's road network traffic space-time state uncertainty calculation method, comprising: collecting historical traffic space-time data;Based on DCRNN model, road network traffic state point prediction model is constructed;Establish the loss function method of describing accidental uncertainty, and using historical traffic space-time data, based on the MC-Dropout of bayesian framework, traffic state uncertainty training is carried out to road network traffic state point prediction model, and the trained road network traffic state interval range prediction model is obtained;The traffic state in the predicted time range is obtained by inputing the traffic space-time data to be predicted into the trained road network traffic state interval range prediction model.The application is more in-depth and effective to quantify the uncertainty of traffic state by combining deep learning technology and bayesian framework, to provide more flexible and comprehensive decision support for traffic management and dispatch.

The application provides a method and system for defect detection and difficult case mining, comprising the following steps: S1. labeled data acquisition; S2. pre-labeling model training; S3. unlabeled data acquisition; S4. initial label generation; S5. artificial review and correction; S6. initial difficult case data generation; S7. reasoning uncertainty calculation; and S8. final difficult case data generation. The application adopts a "defect pre-labeling + artificial review" strategy, only corrects inaccurate defect contours / types, can realize low-cost defect data labeling, can realize multi-class pixel-level defect contour labeling, and solves problems such as missing labeling, slow labeling and poor consistency caused by artificial fatigue.

The invention provides a method for constructing a mathematical analysis model for uncertainty of temperature rise efficiency of a gas compressor. The method comprises the following steps: S1, collecting basic parameters and data; s2, a compressor temperature rise efficiency calculation model is determined; s3, identifying an uncertainty source; s4, performing quantitative calculation on each uncertainty component; s5, calculating the uncertainty of the synthesis standard; s6, evaluating the expansion uncertainty; s7, model verification and feedback correction; and S8, model output and document processing. According to the method, a high-precision and full-data-driven mathematical analysis model for the uncertainty of the temperature rise efficiency of the gas compressor is constructed, the problems of subjective weight deviation, large noise interference and difficulty in compensation of nonlinear interference in traditional uncertainty quantification are solved, and a scientific and reliable uncertainty basis is provided for performance evaluation of the gas compressor.

This invention discloses an intelligent topographic mapping generation method and system based on artificial intelligence, belonging to the field of artificial intelligence technology. It includes real-time data acquisition through a multi-source network integrating satellites, UAVs, and ground sensors, followed by timestamp alignment and noise filtering preprocessing to output a spatiotemporally synchronized data stream. This invention establishes parameterized prior probability distributions for system errors and random noise based on sensor physical models and historical calibration data, accurately classifying measurement errors, environmental noise, and system biases. It configures a lightweight Bayesian model for each data source type and dynamically adjusts uncertainty calculations through environmental impact factors and environmental sensitivity coefficients, achieving environmentally adaptive uncertainty measurement. It converts posterior uncertainty into probabilistic feature weights, directly reflecting data reliability in feature representation, significantly improving the accuracy and reliability of uncertainty quantification.

The invention provides a probabilistic predictive drive self-adaptive high-frequency sampling method and device based on low-frequency flow, and relates to the technical field of power monitoring, and the method comprises the steps: carrying out the low-frequency sampling of a power signal according to a preset first sampling frequency, and obtaining observation data; inputting the observation data into a preset probability prediction model, and outputting an event probability corresponding to the observation data and uncertainty indicating the confidence of the probability prediction model; calculating a second sampling frequency and a buffer window length of high-frequency sampling by using the event probability and the uncertainty; performing power sampling according to the second sampling frequency and the buffer window length to obtain high-resolution waveform data, and performing data processing on the high-resolution waveform data to obtain to-be-uploaded data; according to the embodiment of the invention, the to-be-uploaded data is transmitted to the cloud by using the preset uploading strategy, and the adjustment parameter fed back by the cloud is received, so that the uploaded data volume and the end-side energy consumption can be remarkably reduced on the premise of ensuring the load identification precision and the low omission ratio.

The present invention comprises: a parameter selection unit (101) that selects a parameter; a partial dependency calculation unit (102) that calculates a partial dependency value of an evaluation index value relating to the parameter selected by the parameter selection unit (101), on the basis of a parameter adjustment data set including a plurality of sets of parameter values and evaluation index values, and a trained machine learning model that can output information indicating a predicted value of the evaluation index value; an uncertainty calculation unit (103) that calculates, on the basis of the trained machine learning model and the parameter adjustment data set, an uncertainty value of the evaluation index value relating to the parameter selected by the parameter selection unit (101); and an uncertainty-attached partial dependency plot output unit (104) that outputs information indicating the partial dependency value calculated by the partial dependency calculation unit (102) and the uncertainty value calculated by the uncertainty calculation unit 103.

The present application relates to the technical field of carbon footprint uncertainty analysis, and particularly relates to a carbon footprint uncertainty calculation method and electronic equipment, the method comprising: automatically generating a plurality of rules from unstructured text comprising carbon footprint historical data; fuzzy weighted synthesis of the plurality of rules in a scenario to generate a credibility distribution; extracting a global range and a scenario range according to a credibility threshold, and truncating the probability density function of the obtained truncated normal distribution on the global range to obtain a background distribution; equally dividing the scenario range into three scenario sub-intervals, and calculating the scenario weights using the background distribution; conditionally truncating sampling for each scenario sub-interval, and independently running Monte Carlo simulation to output the corresponding mean and variance; and weighting and synthesizing the Monte Carlo simulation results of each scenario through the corresponding scenario weights to obtain the uncertainty of the carbon footprint. Compared with the prior art, the present application has the advantages of efficient and accurate analysis of carbon footprint uncertainty, etc.

The invention provides a natural gas composition content calculation and corresponding uncertainty evaluation method and system, and belongs to the technical field of natural gas. Calculating the original mole fraction of the sample by adopting a successive normalization method, and performing nonlinear error correction on the original mole fraction to obtain a first mole fraction; sequentially performing data integrity verification, individual component compliance verification and logic correlation analysis of isomeride and carbon number distribution on the original mole fraction to obtain a comprehensive quality evaluation result; and calculating the composition uncertainty of the first mole fraction after the comprehensive quality evaluation result passes. According to the method, accurate, standard, rapid, simple and convenient calculation and evaluation of the natural gas composition content and the uncertainty are realized.

The application discloses a large model text generation method based on semantic adaptation and dynamic hierarchical comparison, and belongs to the technical field of artificial intelligence and natural language processing, which can at least partially solve the problems of lack of semantic perception ability, fixed hierarchical selection strategy and waste of computing resources in the prior art contrast decoding technology, and introduces a semantic routing module in the decoding process.The method obtains the hidden state of different levels of the model and the final layer output distribution when the model reasons and generates each word element; the semantic routing module is used for semantic attribute determination and uncertainty calculation on the current generation state; according to the determination result, a strategy is dynamically selected from a preset decoding strategy set to calculate the final output value; and the final output value is normalized and combined with a sampling algorithm to generate a word element.The application realizes dynamic balance of factuality and fluency, suppresses hallucinations and reduces reasoning delay.

The present invention relates to a system for predicting the progression of a power plant severe accident on the basis of a severe accident analysis database, the system comprising: a power plant state diagnosis module for diagnosing and defining a power plant state and an initiating event that distinguish the progression of a severe accident; a severe accident progression prediction module, which provides first prediction information for predicting the progression of the severe accident on the basis of an accident scenario analysis result extracted from a severe accident analysis database; and a severe accident detailed prediction module, which provides second prediction information for predicting the progression of the severe accident on the basis of an uncertainty calculation result value for each accident scenario stored in a severe accident uncertainty analysis database.