66 results about "Random parameters" patented technology

The invention provides a parallel multi-target random parameter optimization method and device.The parallel multi-target random parameter optimization method comprises the steps that the number of nodes of a parallel computing cluster is configured, and a random seed sequence is generated; carrying out structure loading and compiling on the to-be-optimized model, and obtaining global configuration of the to-be-optimized model; sending the global configuration to all processes, generating an initial parameter population meeting the constraint of the global configuration by each process based on the random seed, and performing serial evaluation to obtain a target function value; independently iteratively executing shuffling complex evolution by each process until a convergence condition is met; each process outputs the respective optimal parameter vector and the corresponding objective function value to a root process, and a global optimal solution is determined through the root process. By means of the method and device, the accuracy and efficiency of parameter optimization can be remarkably improved, and the problem that the model optimization effect is poor in the prior art is solved.

The invention discloses a carbon dioxide geological sequestration quantity prediction method, system and device and a storage medium, and the method comprises the steps: obtaining a mathematical expression of carbon dioxide sequestration capacity through combining a coupling model of a flow field, a mechanical field and a thermal field with random parameters, and introducing a leakage risk coefficient; taking the weighted sum of the expected value of the storage capacity and the variance as an optimization objective function, and according to the mathematical expression of the carbon dioxide storage capacity, carrying out optimal solution solving on the carbon dioxide storage capacity to obtain an optimal storage strategy; performing spatial discretization on the control equation by adopting a finite element method, discretizing time by adopting a semi-implicit difference method, discretizing random variables by combining a random Galkin method, and solving a coupling model of a flow field, a mechanical field and a thermal field to obtain a prediction result of the carbon dioxide sequestration amount; and according to a prediction result of the carbon dioxide sequestration amount, outputting an expected value and a variance of the sequestration capacity, and generating a sequestration amount distribution diagram, a risk assessment diagram and a sensitivity analysis diagram.

The invention provides a glue joint insulation joint fatigue damage evaluation method and system considering random parameters, and relates to the technical field of rail vehicle wheel-rail contact analysis, and the method comprises the steps: obtaining first information; establishing a dynamic contact model of the wheel and the cementing insulation joint according to the first information, and performing simulation to obtain a dynamic response parameter in the contact process of the wheel and the track; performing fatigue damage analysis according to the dynamic response parameters to obtain fatigue damage parameters; performing fatigue life prediction according to the fatigue damage parameter to obtain a fatigue life prediction result; performing sensitivity analysis according to the fatigue life prediction result to obtain a sensitivity analysis result; and calculating the failure probability of the glue joint insulation joint according to the sensitivity analysis result, and evaluating the reliability of the glue joint insulation joint. According to the method, the dynamic contact model of the wheel and the cementing insulation joint is established by using the finite element method and simulation is carried out, so that high-precision wheel-rail contact dynamic response parameters are directly obtained.

The invention discloses a dynamic global sensitivity analysis method for a flap mechanism based on Vine Copula. The method comprises the following steps: firstly, performing rigid-flexible coupling simulation on a movement mechanism, identifying a stress dangerous area, a deformation dangerous area and a clamping stagnation dangerous area of the movement mechanism through a stress distribution condition, and providing a new structure failure criterion according to a dynamic multi-parameter coupling idea, namely a time-varying stress-strain-driving force combined criterion; random parameters are definitely input according to the motion characteristics of the flap mechanism; secondly, establishing a time-varying correlation model considering autocorrelation and cross-correlation based on a Vine Copula theory and a random process model, generating a condition sample by using the model, and generating a required input random variable through inverse transformation; dividing the dynamic sample into a plurality of subintervals by using a space division method, and performing Sobol index solving, so as to solve the importance measure of each input parameter in the movement process of the flap mechanism;

The invention discloses an AI model parameter initialization method based on model parameter and structure multi-modal fusion, and belongs to the technical field of artificial intelligence. The method comprises the following steps: firstly, collecting historical pre-training model data of a cross-model architecture and a cross-data set, processing model parameters into a token sequence, training a Transform codec, converting a model structure into a graph, and training GAT to extract structural features; a multi-modal feature data set is constructed after a related network is frozen, structural features serve as core conditions, a conditional diffusion model DDPM is trained through AdaLN modulation and residual module coupling, and multi-modal feature fusion of'parameter feature-structural feature 'is established. In the reasoning stage, the structural features of the unknown model are extracted, the random parameters of the unknown model are combined with the submerged space shape determined by the encoder, sampling is conducted through the conditional diffusion model, anti-token processing is conducted through the decoder, and adaptive initialized model parameters are generated for the unknown model. According to the method, cross-model structure high-quality parameter initialization is realized, the model training time is shortened, and the large model training requirement is met.

The invention relates to the technical field of cold chain transportation environment control, and discloses a temperature and humidity control system for cold chain transportation, and the system comprises an environment modeling module which is used for constructing a heat conduction model of a cold chain space based on a random field theory and a finite element method, and outputting a stiffness matrix containing random parameters; the self-adaptive sensing network module is used for receiving the characteristic information of the stiffness matrix, dynamically optimizing sensor distribution positions, collecting multi-node temperature and humidity data and adding space-time labels; and the distributed data processing module is used for receiving the sensing data. According to the invention, through random field theoretical modeling and parameter-state joint estimation, the temperature field prediction precision is significantly improved and the system environment adaptability is enhanced; a distributed sensing network architecture and a heterogeneous calculation acceleration technology are combined, so that the cooperative control with optimal energy efficiency is realized while millisecond-level real-time response is ensured, and the performance bottleneck of a traditional cold chain temperature control system is comprehensively broken through.

The application discloses a time-varying reliability topology optimization method for a continuum structure, and comprises the following steps: constructing a time-varying random parameter initial model of the continuum structure based on continuum structure information; constructing a time-varying reliability topology optimization mathematical model according to the time-varying random parameter initial model; obtaining displacement, velocity and acceleration responses of the continuum structure based on the time-varying reliability topology optimization mathematical model; obtaining sensitivity information about random variables according to the displacement, velocity and acceleration responses; obtaining a global time-varying reliability index based on the sensitivity information; establishing a derivative relationship between an objective function and constraint conditions and design variables according to the global time-varying reliability index and an adjoint vector, and obtaining gradient information according to the derivative relationship; and optimizing the continuum structure according to the gradient information. The application realizes high-precision and high-efficiency reliability design of the continuum structure under a dynamic uncertain environment.

The invention relates to a traffic flow speed estimation method based on a Gaussian mixture model and hierarchical Bayesian, and belongs to the technical field of intelligent traffic systems. The method comprises four steps of data preparation and preprocessing, traffic state clustering division, hierarchical Bayesian random parameter model construction, and model verification and performance evaluation. Through deep fusion of the Gaussian mixture model and the hierarchical Bayesian model, data-driven traffic state identification and random parameter estimation are realized, traffic flow heterogeneity can be effectively described, model parameter explosion is avoided, and model prediction precision and generalization ability are improved.

The present application relates to a kind of accident economic loss influence factor analysis method and system, in method part, first based on GB2 model, establish multiple stochastic parameter model, then using LOOIC value carries out effect comparison, selects suitable stochastic parameter model and obtains the positive influence or negative influence of various explanatory variables to accident economic loss;Finally, based on sensitivity analysis, the influence degree of explanatory variable to accident economic loss is excavated.Through the scheme of the present application, an objective, comprehensive, accurate analysis method for excavating accident economic loss influence factor is provided, and the application range is wider, and the influence of related factors on the economic loss caused by accident is better represented.

The invention specifically discloses an ion thruster acceleration grid probability life evaluation method based on coupling competitive failure, and relates to the technical field of aerospace propulsion system life evaluation. The method comprises the following steps: firstly, obtaining deterministic parameters and random parameters of a thruster acceleration grid assembly; then based on an ion sputtering corrosion theory, considering material loss caused by charge exchange ion bombardment of the acceleration gate, and establishing a coupling corrosion model of the thickness and aperture of the acceleration gate; thirdly, constructing a competitive failure model, and realizing numerical solution of competitive failure life by dynamically iterating a corrosion state and a failure criterion; and finally, independently repeated random sampling is carried out based on probability distribution of random parameters, the random sampling is substituted into the competitive failure model to obtain an acceleration grid simulation life sample set, and statistical analysis is carried out on the life sample set to obtain a probability life evaluation result. The method solves the problem of errors caused by independent modeling of structural failure and electronic reflux failure in the prior art, and is helpful for reducing the system operation risk and improving the reliability of a spacecraft.

The application provides a turbine blade thermal barrier coating force-thermal coupling reliability evaluation method, which comprises the following steps: measuring and constructing a density distribution function of a thermal barrier coating microstructure and thermodynamic parameters, completing macroscopic complex turbine blade thermal barrier coating temperature field and strain field simulation, constructing a Monte Carlo sample space of random parameters at different positions of the blade, establishing a micro-scale force-thermal coupling finite element model and obtaining a life data set, building and training a deep learning agent model, and completing turbine blade thermal barrier coating reliability evaluation and sensitivity analysis. The reliability evaluation method provided by the application comprehensively considers the influence of the thermal barrier coating microstructure factor and the force-thermal coupling failure mechanism, solves the problems of high dimension, high nonlinear calculation cost and the like in the reliability evaluation of the complex blade thermal barrier coating, improves the accuracy and efficiency of the reliability evaluation, and provides an evaluation means for the safe application and optimized design of the thermal barrier coating.

This application discloses a method for calculating the thrust envelope of a solid rocket motor based on servo uncertainty, relating to the field of solid rocket motor technology. This method focuses on the gas valve servo system of a throat-operated solid rocket motor, performing uncertainty modeling of the motor drive system and transmission system, analyzing the main influencing parameters of servo system uncertainty, including transmission clearance, eccentric shaft friction coefficient, and other uncertainty parameters and their probability distributions; generating random parameter samples based on the uncertainty model, and using the servo system control model to solve for the actual throat displacement response under random samples, obtaining the engine thrust time history under single-valve and multi-valve cooperative working modes; and fitting the thrust uncertainty envelope by calculating the mean and standard deviation of the thrust response through Monte Carlo simulation. This application quantitatively characterizes the degree of influence of internal factors of the servo mechanism on thrust performance, providing a reference for ensuring precise thrust control of solid rocket motors.

The invention discloses a target export function parameter generation method and device, electronic equipment and a storage medium. The method comprises the following steps: determining an initial parameter structure through an assembly instruction of a target dynamic link library; generating random parameter data based on the initial parameter structure; recording an instruction coverage rate in a process of simulating and executing an export function based on the random parameter data; determining target parameter data of which the instruction coverage rate meets a preset condition; executing an optimization operation based on the target parameter data to determine a parameter value distribution range used for correcting the initial parameter structure; and correcting the initial parameter structure based on the parameter value distribution range, and generating a target derivation function parameter. A dynamic execution process combining static analysis and optimization operation is realized, the problems that effective input is difficult to generate and the instruction coverage rate is low in unsigned information DLL parameter identification are solved, and efficient and accurate identification of the derived function parameter structure is realized.

The invention relates to a stochastic simulation method considering dimensional deviation and concrete material variability, and belongs to the field of concrete structure analysis. According to the method, accurate simulation of the mechanical properties of the concrete structure under the influence of construction errors and material randomness is realized by determining random factors, establishing a random field matrix, carrying out sample transformation, establishing random models in batches, solving in batches and carrying out post-processing. A random simulation result is used as an output parameter, a random sample is used as an input parameter, and a PRRM-EPR algorithm is adopted to establish a machine learning prediction model. Compared with a traditional method, the method has the advantages that the generation quality of the random sample is improved, the calculation cost is reduced, the randomness in actual engineering can be better reflected, and the established prediction model can directly predict the mechanical response (displacement, load and the like) of the component through actually measured random parameters. The method is suitable for design, evaluation and optimization of a concrete structure, and particularly has a remarkable effect on improvement of structural safety.

The invention discloses a time-varying thermal failure probability estimation method and system for a deep groove ball bearing, and belongs to the field of bearing design. The method comprises the following steps: constructing a basic parameter model containing a mixed random parameter vector and a limit state function; establishing a transient thermal network model to calculate time-varying temperature response; constructing an agent model to replace high-time-consumption heat generation rate calculation; a self-adaptive time node sampling strategy is adopted, a proxy model and a thermal network model are combined to process large-scale samples, and the time-varying thermal failure probability is estimated. The system comprises a basic parameter modeling module, a transient thermal network simulation module, an agent model construction module and a self-adaptive probability evaluation module which correspond to the system. According to the method, the time-varying and random dual uncertainty is fused, and the proxy model and the adaptive sampling strategy are utilized, so that the defects of conservative property of a traditional static evaluation method and low efficiency of a traditional probability simulation method are effectively overcome, and accurate evaluation of the thermal behavior of the bearing under the dynamic working condition and remarkable improvement of probability calculation efficiency are realized.

The invention provides a geological analysis parameter protection method and system based on logarithm mapping, and relates to the technical field of geological exploration data protection, and the method comprises the steps: obtaining an original R parameter, building a linear mapping relation g = aR + b, building logarithm transformation mapping based on the linear mapping relation, converting the original parameter into a transformed parameter value, and obtaining a parameter value; and the transformed parameter values are provided for the data processing party for geological exploration operation, and the data owner recovers the original parameters through an inverse transformation formula after receiving the data obtained by exploration of the data processing party. According to the method, through logarithm transformation, an exploration company cannot identify the stratum type through conventional linear fitting, and sensitive geological information is effectively protected; meanwhile, irreversibility of the transformation process is ensured by adopting random parameters, and a data owner can completely recover original geological parameters through an inverse transformation formula; on the premise of protecting the data security, the normal exploration operation flow and the data processing precision are not influenced.

The invention relates to the technical field of signal processing, in particular to a filtering method for dealing with a system with complex multiplicative noise and random time delay uncertainty, which designs a reasonable system model of a multi-sensor networked system meeting conditions, and adopts a maximum and minimum robust estimation principle, a de-randomization method and a virtual noise technology. The system is converted into a multi-model multi-sensor system only with uncertain noise variance, an augmented noise method, a non-negative definite matrix factorization method and a Lyapunov equation method are applied, two robust centralized fusion steady-state Kalman estimators are obtained, and the robustness of the estimators is further proved. The method can be used for solving the problem of robust fusion Kalman filtering of multi-sensor single-channel ARMA signals with random parameter matrixes, uncertain noise variance and networked random uncertainty.

This invention belongs to the field of power system demand-side management technology and discloses a probabilistic assessment method for load adjustability potential considering dual uncertainties. The method includes: constructing adjustability potential models for five load types: air conditioning, cold chain, energy storage, electric vehicles, and data centers; determining the random parameters of external influences on each load and quantifying their probability distribution based on historical data; introducing random execution variables for each load type to describe the uncertainty of execution behavior with probability distributions, thus obtaining a more realistic adjustability potential; performing random sampling calculations on the above models and parameters using the Monte Carlo method to aggregate and generate a total adjustability potential sample; and performing fitting analysis on the sample to obtain the probability function of the system's adjustability potential and conduct multi-dimensional evaluation. This invention achieves the quantification of load adjustability potential uncertainty through a three-layer probabilistic modeling of "physical behavior - random influence - random response," providing decision support for optimized power grid operation.

The invention provides a power distribution network planning method based on a Monte Carlo model, and belongs to the technical field of power distribution network planning. In order to solve the technical problem that a decision maker cannot make a more accurate power distribution network planning scheme on the basis of potential benefits brought by various technical development paths at present, the adopted technical scheme is as follows: defining targets and constraint conditions of power distribution network planning as follows: based on constraint conditions of voltage deviation, line capacity, short-circuit current and an N-1 criterion, determining a power distribution network planning model; the planning targets of minimizing the investment cost, maximizing the reliability and optimizing the renewable energy source permeability are achieved; establishing a probability model for the key uncertainty factors by adopting a Monte Carlo model based on probability statistics and stochastic simulation, generating various possible load and renewable energy output scenes in the future through the Monte Carlo model, and generating a random scene combined by multiple groups of random parameters through probability distribution; the method is applied to power distribution network planning.

The application relates to a traffic flow speed estimation method based on a Gaussian mixture model and a hierarchical Bayesian method, and belongs to the technical field of intelligent transportation systems. The method comprises four steps of data preparation and preprocessing, traffic state clustering division, hierarchical Bayesian random parameter model construction, model verification and performance evaluation. Through the deep fusion of the Gaussian mixture model and the hierarchical Bayesian model, the application realizes data-driven traffic state recognition and random parameter estimation, can effectively depict traffic flow heterogeneity, avoids model parameter explosion, and improves the model prediction precision and generalization ability.

The application belongs to the technical field of structure optimization, and particularly relates to a stiffened plate structure reliability topology and size joint optimization method, which comprises the following steps: taking a three-dimensional stiffened plate structure as an object, selecting the mean value of each random parameter as the initial value, and performing deterministic topology optimization; performing structure reliability analysis based on an HMV algorithm, and searching for the current most likely failure point; judging whether the target converges or not, regularizing and simplifying the material layout of the topology optimization output structure; selecting the mean value of each random parameter as the initial value of the normalized steel structure of the layout explanation, and performing deterministic size optimization; performing structure reliability analysis based on the HMV algorithm, and searching for the current most likely failure point; judging whether the target converges or not, and obtaining the stiffened plate optimization structure which simultaneously satisfies the structure performance and reliability index. The application realizes the reliability topology and size joint optimization of the material distribution optimization and the production demand satisfaction while ensuring the reliability.

The present invention discloses a seafloor inversion method, apparatus, device, and medium based on genetic clonal selection. The method comprises: initializing a first population and using the first population as a target population; performing inversion and evaluation based on inversion parameters of the target population to obtain the root mean square value corresponding to each individual; setting random parameters based on the minimum root mean square value and the number of population iterations; updating the population by improving a genetic algorithm based on the random parameters; when an individual in the updated population has a root mean square value that meets an optimal iteration threshold, using the corresponding individual as the target individual; otherwise, cyclically updating the population until the target individual is obtained or the number of population iterations equals a preset maximum number of population iterations; and finally determining the seafloor inversion result of the area to be inverted based on the seafloor topography data corresponding to the target individual. The present invention can accurately achieve seafloor inversion and can be widely applied in the field of data processing technology.

The invention discloses a bird repelling method, system and device based on multi-mode intelligent perception and collaborative repelling and a storage medium, and relates to the technical field of electric power facility protection and intelligent bird repelling, the method comprises the steps of continuously scanning a monitoring airspace, and constructing a standardized dynamic parameter sequence; calling a behavior prediction model to calculate the probability that the birds approach the intention, comparing the probability with a preset threshold value, and generating a bird repelling starting instruction when the probability exceeds the threshold value; if not, the low-power-consumption monitoring state is maintained; after the bird repelling starting instruction is triggered, a random algorithm is triggered, and a group of random parameters for uniformly controlling ultrasonic, sound, light and laser modules are generated; the parameters are analyzed, all the modules are synchronously driven to cooperatively work in a random combination mode for a period of random time, and the system automatically returns to a monitoring state after bird repelling is completed; according to the invention, intelligent sensing and multi-mode random collaborative driving of bird approaching behaviors are realized, the bird driving durability can be obviously improved, and safe and stable operation of electric power facilities is guaranteed.

This invention discloses a smart home energy management method based on deep adversarial inverse reinforcement learning, comprising the following steps: (1) Modeling the energy cost minimization problem of a smart home and designing the corresponding Markov decision process environment state and actions; (2) Constructing n sets of random parameter historical data based on a set of random parameter historical data and a rolling generation method; (3) Solving the above minimization problem using the n sets of random parameter historical data and an optimization algorithm to obtain n expert trajectories; (4) The discriminator trains a reward network based on the expert trajectories and the trajectories generated by the generator agent; Under the guidance of the reward network, the generator agent is trained using experience tuples and a proximal policy optimization algorithm; (5) Repeating step (4) until a stable agent policy is obtained; (6) Deploying the trained agent policy in a real environment. Compared with existing methods, the method of this invention can effectively reduce energy costs and improve user comfort.

The invention relates to the technical field of chip simulation verification, and discloses a Macro IP parameter verification method, device and equipment, and the method comprises the steps: obtaining a parameter identifier for generating a Macro IP; based on the verification target, judging whether a parameter represented by the parameter identifier is a random parameter, and if the parameter represented by the parameter identifier is judged to be the random parameter, constraining a parameter value of the random parameter to obtain a target random parameter and a random parameter expected value corresponding to the target random parameter; if the parameter represented by the parameter identifier is judged to be the preset parameter, modifying a parameter value corresponding to the preset parameter to obtain a target preset parameter and a preset parameter expected value corresponding to the target preset parameter; generating a register-level transmission code based on the target random parameter and the target preset parameter; and constructing a test case, and verifying whether the operation result of the register-level transmission code meets the expectation or not. Manual parameter modification can be replaced, automatic verification is realized, and verification efficiency is improved.

The application discloses a watershed runoff probability prediction method based on Bayesian deep learning and a differentiable physical model, and the method comprises the following steps: firstly, a digital elevation model is used to construct a watershed differentiable distributed computing architecture, a parameter inference network is connected with an HBV hydrological model and a Muskingum confluence equation in series to form an end-to-end differentiable computation graph; then, Bayesian inference is realized by introducing a random inactivation layer and an input disturbance mechanism, and a random parameter group and a random rainfall scenario conforming to a posterior distribution are generated; finally, parallel set physical evolution is performed, and a runoff probability prediction and a flood risk assessment interval are output; the method combines a physical mechanism and deep learning, realizes high-precision deterministic prediction and reasonable uncertainty quantification, and provides more reliable decision support for flood warning.

The application belongs to the field of data processing and relates to an improved estimation method for distributed random parameters under the condition of probability density overlap, comprising the following steps: 1) obtaining the dynamic parameters of an engineering structure; the dynamic parameters include the mass, the moment of inertia and the running speed of the structure; 2) performing distributed random parameter estimation on the dynamic parameters obtained in step 1), so as to improve the accuracy of the probability density of the dynamic parameters under the condition of probability density overlap; and 3) performing stability analysis on the engineering structure based on the result obtained in step 2); the stability analysis of the engineering structure is performed by predicting or evaluating the running state of the engineering structure. The application provides an improved estimation method for distributed random parameters under the condition of probability density overlap, which can reduce the estimation error and improve the accuracy of data processing.

The present application provides a method, device, equipment and storage medium for reducing the risk of sticking, comprising: obtaining input data of a prediction model and sticking probability corresponding to the input data, wherein the input data comprises fixed parameters and controllable parameters; when the sticking probability is higher than a first sticking threshold, generating a random population according to the controllable parameters; after adaptive screening, crossover operation and mutation operation on the random population, obtaining random parameters; inputting the random parameters and the fixed parameters into the prediction model to obtain an optimized sticking probability; if the optimized sticking probability is lower than a second sticking threshold, using the random parameters as the controllable parameters to guide the deployment of horizontal well drilling; when the sticking probability is higher than the first sticking threshold, generating random parameters according to the controllable parameters and inputting the random parameters and the fixed parameters into the prediction model to obtain an adjusted sticking probability according to the prediction model; when the sticking probability is lower than the second sticking threshold, using the random parameters to guide the deployment of horizontal well drilling.

The invention discloses a bridge fatigue reliability analysis method, system and device and a medium, and relates to the technical field of structural reliability analys.The method comprises the steps that related structural parameters in a bridge fatigue failure performance function are obtained, and a Copula model is adopted to construct a joint probability density function of a high-dimensional random variable; random variables of original space distribution are mapped to an independent standard normal distribution space for independent sampling, independent standard normal random samples are generated, and the independent standard normal random samples are reflected into original space distribution to obtain independent random samples of joint probability density function input random variables; calculating a plurality of performance function response values by adopting an independent random sample, selecting the plurality of performance function response values as a threshold value of an intermediate failure event, counting the number of newly generated samples falling into a failure domain, and calculating to obtain a structure failure probability; the method solves the problem that a traditional method neglects the dependency relationship between random parameter variables, and consequently the accuracy of a bridge fatigue reliability analysis result is low.

The invention relates to the technical field of power system control and stability analysis, in particular to a quantitative analysis method and device for control performance by source network uncertainty and a medium, and the method comprises the steps: generating a plurality of random parameter combinations based on a plurality of probability distribution models; automatically configuring the random parameter combination to a pre-established power system simulation model, driving the power system simulation model to complete simulation operation of corresponding times, and automatically extracting at least one controlled quantity data; and calculating at least two performance indexes on the basis of the extracted controlled quantity data, and establishing a mapping relation between the random parameter combination and the performance indexes on the basis of the data of each simulation frequency so as to quantify the influence degree of different uncertainty factors on each control performance index. Through the setting, the influence of the bilateral uncertainty of the source network on the control performance of the power system can be systematically quantified, and the problem that the coupling effect of the bilateral uncertainty cannot be comprehensively analyzed by a traditional method is effectively solved.