8 results about "Interval estimation" patented technology

The application discloses a natural gas load interval estimation method based on large model coupling working condition clustering, and belongs to the technical field of natural gas pipeline network operation optimization and artificial intelligence load prediction. The method extracts working condition semantic constraints by using a large model, and obtains fuzzy working condition clusters by combining historical operation data clustering; a natural gas load point prediction model is trained for each cluster, and residual probability density distribution is estimated; semantic and numerical weights are fused in real time, and the final point prediction value and the natural gas load prediction interval at the prediction time are obtained by dynamic weighting, which are taken as the natural gas load prediction result and output. The application introduces a large model into the natural gas working condition expression and clustering constraint construction process, and no longer uses the large model as a simple downstream feature generation tool, but solves the problem of interval estimation failure of the natural gas load under fuzzy working conditions such as holiday switching, peak-valley transition and extreme weather through deep coupling of the large model and working condition clustering, so that high-reliability dynamic prediction interval output under complex working conditions is realized.

The application discloses a deep-cut valley stress field inversion method based on a neural network and a numerical model, and comprises the following steps: deep-cut valley stress field inversion region determination and numerical model construction; stress interval estimation, stress boundary parameter sample generation; batch forward calculation, measurement point stress-boundary stress data set construction; BP neural network construction for inversely deducing stress boundary conditions from geostress measurement point data; actual stress boundary conditions are inversely deduced from the measured geostress of the measurement points, and the final geostress field is calculated by combining the physical information neural network and the numerical model in a forward direction. The application embeds physical constraints such as balance differential equations and compatible equations into the loss function of the physical information neural network, solves the problem that the existing method is biased towards data statistics and the physical mechanism is unclear, realizes the physical rationality of the inversion result, generates parameter samples through orthogonal test design and carries out batch numerical forward calculation, and fully covers the parameter space under the condition that the measurement point data is sparse and the distribution is limited.

This invention relates to the field of product reliability analysis technology, specifically to a reliability assessment method for a non-fixed shape parameter dependent competing failure model. The method includes the following steps: constructing a dependent competing failure model, designing an adaptive stepwise type II hybrid truncated constant-load life test, estimating parameter points based on test data, establishing an acceleration equation, calculating product reliability under normal stress, and performing interval estimation. This invention constructs a MOBG distribution description that assumes fixed shape parameters and does not consider the dependencies between unknown failure mechanisms and causes. It designs an adaptive stepwise type II hybrid truncated constant-load life test to obtain data, solves for parameters based on maximum likelihood estimation and Bayesian estimation, and combines the acceleration equation to obtain parameter estimates under normal stress. This leads to the calculation of product reliability and interval estimation, improving the accuracy and applicability of reliability assessment. It is suitable for reliability analysis of various products with multiple failure mechanisms and potentially unknown causes.

The application provides a power electronic device inertia and primary frequency modulation coefficient estimation method, comprising: collecting unit, regional and system real-time data; inputting the unit, regional and system real-time data into a trained collaborative adaptive counter particle filtering module, and outputting a posterior particle set; the collaborative adaptive counter particle filtering module comprises an agent network and a particle transformation network; performing statistical operation on the posterior particle set to obtain a power electronic device inertia and primary frequency modulation coefficient estimation value. The method provided by the application can realize long-term, adaptive, point estimation and interval estimation of inertia and primary frequency modulation coefficient in a multi-level unit, region and system, solve the problems of model dependence and particle degradation of existing methods, and improve the robustness and engineering applicability of power electronic device dynamic parameter evaluation.

This invention discloses a fault detection method and system for Mecanum wheel automated guided vehicles (AGVs), relating to the field of industrial automation fault diagnosis technology. Based on differential inclusion theory and the Euler-Lagrange equations, this invention establishes a Lur'e differential inclusion-type state-space equation characterizing system nonlinearity, parameter uncertainty, and external disturbances. An adaptive event-triggered mechanism with dynamically adjustable parameters and exponentially decaying terms is designed to optimize data transmission, reduce resource consumption, and strictly exclude Zeno behavior. An adaptive event-triggered interval observer is constructed to achieve robust interval estimation of the system state, and actuator fault detection is achieved through the zero-value inclusion property of the residual interval. This invention solves the problem of traditional methods struggling to balance fault detection accuracy and resource utilization, and is suitable for state monitoring and fault diagnosis of Mecanum wheel AGVs in industrial scenarios.

The application provides a load identification and structure response reconstruction method based on denoising regularization, comprising the following steps: step one, deducing reconstruction equations of structure dynamic response and external load based on a state space model; step two, introducing a wavelet transform threshold method to perform denoising processing on the measured response; step three, performing interval estimation on the denoised measured response by using a grey mathematical method to determine the interval median and radius of the response; step four, analyzing the ill-posed problem in load identification by means of Tikhonov regularization, so as to inverse the estimation interval of the external load; and step five, realizing interval response reconstruction of the structure by using the dynamic response reconstruction equation. The application has higher effectiveness and applicability, and can quickly and effectively identify the uncertain interval of the load and the response.

The present application is directed to the vulnerability of B5G industrial internet system and the attack threat problem, based on the current existing network security measurement system, which is more suitable for traditional internet, ignoring the current situation of B5G industrial internet and its differences, a system design for realizing B5G industrial internet network security measurement is provided, and the technical scheme is: firstly, based on the risk analysis of industrial network environment to the information system, to build attack / threat sample library, at the same time, combined with point estimation, confidence interval estimation, B5G industrial internet network security event hit rate, security degree, system security health coefficient calculation formula, build multi-dimensional security measurement comprehensive coordinate analysis system. The present application provides a multi-dimensional B5G industrial internet security measurement system, which can assist in improving the security of B5G industrial internet, and provides protection for efficient and safe production.

This application discloses a method and related equipment for predicting the functional failure probability of shield tunnels based on multiple variables. The method includes: constructing an integrated damage-function index system covering multiple scales and quantifying the mapping relationship between damage and function; optimizing the probabilistic seismic demand model using interval estimation methods and selecting the best vibration intensity index by combining fuzzy multi-criteria decision-making; introducing the Copula function to accurately characterize the nonlinear correlation between multidimensional damage / function parameters; and finally establishing a multivariate vulnerability analysis model that can output the failure probability of each function of the tunnel and the overall failure probability of the system. This provides reliable technical support for the seismic design, post-earthquake assessment, and operation and maintenance decisions of shield tunnels. It can accurately assess the seismic risk of shield tunnels under non-uniform liquefaction development, improve the reliability and robustness of the results, and can be widely applied in the field of computer technology.