177 results about "Random model" patented technology

The invention provides a method for participating in primary frequency modulation control of a power system based on a modular large-scale hydrogen production power supply, and the method comprises the steps: obtaining an original data flow from a high-precision synchronous phasor measurement unit, an electrolytic cell and a submodule, and carrying out the multi-scale decomposition and noise reduction processing after time synchronization and anomaly detection restoration; performing state estimation by using unscented Kalman filtering, extracting frequency disturbance characteristics in combination with empirical mode decomposition and Hilbert transform, and estimating the inertia of the power grid; an electrolytic cell dynamic model is established, and parameter identification and optimization are carried out through a recursive least square method; a control strategy is designed based on a multi-time scale control framework, virtual synchronous generator parameters are optimized through online Gaussian process regression, and frequency modulation control is realized in combination with scene tree random model predictive control and minimum-maximum regret learning; and finally, carrying out coordination control through a distributed consistency algorithm and carrying out multi-target performance evaluation optimization. And the frequency modulation performance and the operation efficiency of the system are improved.

The invention relates to the technical field of underwater intelligent navigation and integrated navigation, in particular to a submersible integrated navigation multi-source data fusion method based on an elastic random model, which comprises the following steps: preprocessing multi-source navigation data output by an inertial navigation system (INS), a Doppler velocimeter, an ultra-short baseline / long baseline, an altimeter and a depthometer; constructing a state space model containing navigation parameter errors and sensor errors, and screening an optimal sensor combination mode; the method comprises the following steps: constructing an elastic correction parameter model and a dynamic adjustment state and observation noise covariance matrix by combining an elastic PNT framework, performing local filtering on multi-source data by adopting a plurality of parallel elastic self-adaptive robust volume Kalman filters (CKF), outputting a fusion resolving result, and realizing optimal estimation of a global navigation state through a dynamic information distribution strategy. The method has relatively high anti-interference performance, self-adaptive capability and combination switching robustness, and is suitable for navigation and positioning tasks of a high-precision submersible in a complex underwater environment.

The present disclosure discloses a three-dimensional inversion method of airborne transient electromagnetics based on deep learning. The method of the present disclosure proposes two strategies that focus on training datasets, to improve the performance of deep learning models, including divide and conquer strategy and random models generating. Through a large of reasonable structural models, appropriate network setups, a more generalized result can be obtained through our proposed U-Net framework, which has been demonstrated to be effective on both synthetic and field data. This scheme can realize the rapid prospecting of three-dimensional resistivity structure in large-area target region, and solve the problem of low efficiency of traditional three-dimensional inversion calculation of ATEM and poor migration ability of three-dimensional inversion based on deep learning developed by predecessors.

The invention provides a rapid random model correction method for an active learning type ocean platform structure, and the method comprises the steps: firstly carrying out the selection of correction parameters through considering the sensitivity, then carrying out the design of a test scheme, and building an initial Kriging model; a new correction parameter training sample is found, the frequency corresponding to the new sample is calculated, a Kriging model is reconstructed by adopting a new training set, random model correction is performed based on the reconstructed Kriging model, an objective function is constructed by adopting actually measured frequency data and initial finite element model frequency data, and a correction parameter mean value and a standard deviation are taken as solving objectives. And solving by utilizing an optimization algorithm to obtain a mean value and a standard deviation of correction parameters of the finite element model to be corrected, and repeatedly iterating to realize random correction of the ocean platform structure. According to the method, rapid, efficient and accurate model correction under small samples and noise errors is realized, and higher correction accuracy can be obtained by adopting fewer training samples.

The invention relates to a point cloud registration method for constructing a random model based on a synthetic variance and covariance matrix. The method comprises the following steps: S1, obtaining a coarse registration transformation matrix and a corresponding coarse registration transformation point cloud; s2, calculating distances between a query point and all points in the KD-Tree index structure, and determining the number of nearest neighbor points corresponding to the query point based on the distances; s3, screening a new corresponding point pair set; s4, solving a cofactor matrix; s5, calculating an estimated value of a fine distribution conversion parameter and an estimated value of a random error; s6, calculating source point cloud coordinates in a new state, and if the source point cloud coordinates are smaller than a set threshold value, integrating the fine registration transformation parameters after iteration and the coarse registration transformation matrix to obtain a final transformation parameter matrix; and otherwise, taking the source point cloud coordinates in the new state as the current point cloud data, taking the conversion parameter estimation value as the current fine registration parameter, and returning to S1. Compared with the prior art, the method has the advantages of realizing higher-precision parameter estimation and the like.

The invention discloses an autonomous region division method based on random model prediction control under opportunity constraint. The method comprises the following steps: predicting wind power photovoltaic output by using a differential autoregressive moving average model; considering the uncertainty of wind and light output, and establishing an autonomous region division opportunity constraint system by taking the minimum load loss risk rate of the autonomous region as a target; generating a distributed output scene set in a specified reasonable output interval by adopting a scene analysis method in combination with photovoltaic and fan prediction error probability models, and solving opportunity constraints by applying a stochastic simulation technology; a novel power distribution system autonomous region division optimization model based on random model prediction control is constructed, and continuous forward rolling optimization is carried out through rolling optimization and feedback correction. The method is used for solving the problem that in the prior art, the control dimension is too high due to the fact that distributed energy is connected into a distribution network, it can be guaranteed that the load loss risk probability is kept at a low level under a certain confidence level, and consumption of distributed new energy and the stability of an autonomous region are improved while the power supply reliability is guaranteed.

The invention discloses a wind-light-diesel-V2G micro-grid active frequency random model prediction control method, and belongs to the technical field of micro-grid frequency control. The invention aims to provide a wind-light-diesel-V2G micro-grid active frequency random model prediction control method for solving a micro-grid frequency control problem by combining an active frequency random model prediction secondary frequency control method of an unknown input decoupling observer. The method comprises the following steps: establishing a charging / discharging power model considering the charging demand of an EV user, establishing a micro-grid model containing a wind-light-diesel charging pile, establishing an unknown input observer of an improved Luenberger observer, and predicting and controlling secondary frequency optimization control based on a scene tree random model. According to the invention, the control effect under the uncertain wind-light load condition is improved. The micro-grid frequency control effect of the micro-grid under the uncertain conditions of wind, light and charging piles is improved, and large-scale replacement of distributed wind and light new energy and further popularization of electric vehicles are facilitated.

The invention relates to a GNSS (Global Navigation Satellite System)-based water equipment positioning monitoring method, device and system, and the method comprises the steps: eliminating an error in an observation value caused by natural signal interference through building a mixed elevation angle and signal-to-noise ratio random model which gives consideration to an elevation angle and a signal-to-noise ratio; multi-path errors in observed values caused by water surface reflection are eliminated by establishing a multi-path error correction model, and meanwhile, natural signal interference and water surface multi-path error influence are further eliminated through error correction in the long baseline resolving process by adopting a GNSS long baseline resolving algorithm considering the multi-path errors, so that the real-time performance of the GNSS long baseline resolving algorithm is improved. High-precision stable calculation of a large-area water area is guaranteed, and the problem that in the existing position monitoring process of the water equipment, due to water surface fluctuation and multi-path signal interference, the detection result is inaccurate is solved.

Described herein are systems, methods, and other techniques for training and deploying a random forest model to predict a presence of ASD in a subject. A training dataset can be obtained including abundance measurements of a first cohort of subjects with ASD and a second cohort of subjects without ASD across the set of microbial markers. Multiple candidate random forest models are trained by generating a validation dataset comprising one of k subsets of the training dataset and a training subset comprising the remaining k-1 subsets, training a candidate random forest model of the k candidate random forest models using the training subset, and assessing the candidate random forest model using the validation dataset by computing performance metrics for the candidate random forest model. Candidate random forest models are evaluated by computing an AUC.

The invention discloses an ionospheric scintillation monitoring method based on a geodetic measurement receiver, which belongs to the technical field of satellite navigation, and comprises the following steps of: accurately eliminating non-scintillation errors by constructing an inter-station single-difference function model and an inter-station single-difference random model, extracting carrier phase residual errors by using a single epoch ambiguity fixing technology, and finally obtaining an ionospheric scintillation monitoring result. Establishing a reliable ionospheric scintillation monitoring index; according to the ionospheric scintillation monitoring method based on the geodetic measurement type receiver, high-precision real-time monitoring of ionospheric scintillation can be realized through a single-epoch ambiguity fixing technology under the condition of not depending on an external precision product, and reliable technical support is provided for space weather monitoring.

The invention discloses a gravitational field inversion method considering unsteady noise of a satellite gravity observation value. The gravitational field inversion method comprises the following steps: A, acquiring original observation data of a gravity satellite; b, setting observation value noise as white noise, and constructing a random model considering non-stationary noise; c, performing parameter estimation to obtain a post-test residual error of an observation value; d, estimating the noise auto-covariance and precision factor matrix of the observation value based on the post-test residual error of the observation value; e, repeating the steps C and D for iteration to realize refinement of the gravitational field parameter and observation value random model; and F, evaluating the time-varying gravity field model in the spectral domain. According to the invention, the defects in the prior art can be improved, and the precision and stability of gravitational field inversion are improved.

The invention relates to the technical field of wireless and terminals, and particularly provides a positioning method and device, electronic equipment and a computer program product. The method comprises the following steps: acquiring a first ranging signal of a first positioning source and a second ranging signal of a second positioning source corresponding to a to-be-positioned target; the observation noise is determined based on the first ranging signal, the second ranging signal and a preset random model, and the preset random model is determined at least based on a first signal characteristic of the first observation noise and a second signal characteristic of the second observation noise; determining a positioning combination result based on speed constraint and observation noise; and performing nonlinear processing on the positioning combination result, and determining a final positioning result of the to-be-positioned target. According to the invention, the random model is refined according to the characteristics of different positioning sources, the continuous smooth positioning of the transition scene is achieved, and the linearization error is weakened, so that the availability, smoothness and accuracy of the positioning result are improved.

The invention provides a construction method of a fiber-matrix bond slip random model, and belongs to the technical field of composite materials. The method comprises the following steps: designing fiber burial depth and pull-out angle parameters, preparing a bidirectional single fiber pull-out test piece, carrying out a pull-out test to obtain load-displacement data, drawing a curve, analyzing characteristic value data of the curve, simplifying a bond-slip relationship into a trilinear model, introducing Weibull random distribution, and carrying out parameter estimation and inspection on the Weibull random distribution. And finally establishing a function relationship between the Weibull random distribution parameter and the fiber burial depth and the pull-out angle. According to the construction method of the fiber-matrix bond slip random model, the constructed model can generate a unique single fiber pull-out curve according to the characteristics of each fiber in the composite material, and the problem that the single fiber pull-out randomness is difficult to describe by a conventional theoretical model is solved.

The invention discloses a time-varying channel estimation method and device based on correlation prediction in an RIS auxiliary system, and the method comprises the steps: assuming that the working mode of the RIS is an ON or OFF mode, and obtaining time domain cascade channel estimation at a historical moment at a base station side; constructing a training sample set by using time domain cascade channel estimation at historical moments; randomly initializing network parameters, and training the long and short-term memory neural network by using the training sample set to obtain a network model with an optimal weight and threshold; predicting the time correlation of the channel based on the trained network model; generating an AR random model by using the time correlation of the channel, and constructing a state space model of the Kalman filter based on the AR random model and the time domain receiving signal; time-varying channel estimation is carried out based on the state space model of the Kalman filter, and cascade channel estimation passing through each RIS unit is obtained. The method is suitable for high-precision acquisition of the time-varying channel information in the RIS-assisted mobile communication system.

The invention provides a vehicle-mounted high-precision positioning method based on virtual 3D speed adaptive constraint, and relates to the technical field of vehicle navigation. The method comprises the following steps: under the condition that a navigation system GNSS (Global Navigation Satellite System) signal of a vehicle is interrupted, acquiring measurement data of an inertial measurement unit in the vehicle navigation system, constructing a regression model of a virtual 3D speed of the vehicle, establishing a mapping relation between the measurement data and the virtual 3D speed of the vehicle, and generating the virtual 3D speed of the vehicle; constructing an observation model according to the virtual 3D speed, and adaptively optimizing a random model by evaluating the regression precision of the virtual 3D speed; and determining the positioning information of the vehicle according to the observation model and the random model. According to the method, the accuracy of the positioning information provided for the vehicle can be improved under the condition that GNSS positioning is unavailable due to a complex environment.

The invention discloses a digital twinborn emergency rescue deduction optimization method for a complex disaster scene, and the method comprises the steps: constructing a composite random model for emergency resources, rescue task demands and the high dynamics and randomness of environmental elements in the disaster scene through employing a random process theory and a discretization method, and carrying out the deduction optimization of the digital twinborn emergency rescue. A deduction algorithm of a continuous state space and discrete time step length is constructed based on a digital twin deduction technology, bidirectional mapping of a physical space and a virtual space is realized, an emergency rescue optimization module based on a matching game mechanism is embedded in a digital twin deduction framework, a task demand gap and resource cost are taken as objective functions, and the task demand gap and the resource cost are taken as objective functions. And a global optimal solution is gradually approached by utilizing multi-round iteration and local strategy exchange, so that efficient optimization of resource allocation in a disaster scene is realized. According to the method, the future disaster trend can be predicted in real time, and dual optimization of task demand satisfaction and resource cost control can be realized.

The invention relates to a throughput optimization method and device and an unmanned aerial vehicle control system. The method comprises the following steps: acquiring a signal to interference plus noise ratio of a target unmanned aerial vehicle; the target unmanned aerial vehicle comprises an unmanned aerial vehicle in a U2D mode; obtaining the throughput of the unmanned aerial vehicle in the U2D mode according to the signal to interference plus noise ratio of the unmanned aerial vehicle in the U2D mode; under a preset constraint condition, according to the throughput, constructing a random model for optimizing the total throughput of all the unmanned aerial vehicles in the U2D mode; performing iteration based on the random model to obtain an unmanned aerial vehicle operation strategy; the unmanned aerial vehicle operation strategy comprises a strategy that the target unmanned aerial vehicle carries out resource allocation and trajectory planning to enable the total throughput to be optimal. According to the method, the random model for optimizing the total throughput can be constructed according to the throughput of the unmanned aerial vehicle in the U2D mode, and the unmanned aerial vehicle operation strategy capable of enabling the total throughput to be optimal is obtained through iterative optimization based on the random model.

The invention relates to the technical field of ore raw material analysis, and discloses an intermediate fusion strategy ore sample classification method based on random model combined data, and the method comprises the following steps: S1, collecting LIBS spectrum and Raman spectrum data of a manganese ore sample; s2, preprocessing the spectral data; s3, performing dimensionality reduction on the preprocessed LIBS and Raman data by adopting principal component analysis; s4, splicing the feature matrixes after dimension reduction to construct an intermediate fusion data set; s5, training a random forest classification model based on the intermediate fusion data set; and S6, carrying out classification and identification on the manganese ore samples by utilizing the trained model. According to the method, metal element characteristic spectral lines captured by the LIBS spectrum and molecular vibration characteristics identified by the Raman spectrum are organically integrated through an intermediate fusion strategy, two-dimensional identification is achieved, redundant information is effectively eliminated while main characteristics of the spectrum are reserved, and the accuracy and reliability of manganese ore sample classification are remarkably improved.

The invention discloses a multi-energy micro-grid energy regulation and control method and device and a storage medium. The method comprises the following steps of: constructing a scene prediction constraint optimization problem by utilizing a trained Warisstein generative adversarial network, and predicting a group of multi-dimensional uncertainty multi-energy micro-grid operation scenes in a future period by following known information at each moment to serve as prediction scenes; forming a pre-scheduling model and a real-time power compensation model based on the prediction scene, and forming a two-stage multi-energy micro-grid real-time energy regulation and control model; a pre-scheduling model obtains a pre-scheduling instruction of the multi-energy micro-grid based on random model prediction control by using the prediction scene information; the real-time power compensation model coordinates schedulable units in the multi-energy micro-grid according to real-time observation of uncertain variables; therefore, the energy in the multi-energy micro-grid is regulated and controlled. Through quick response and adjustment of real-time data, the system can effectively cope with fluctuation caused by uncertainty, thereby realizing accurate regulation and control of the energy of the multi-energy micro-grid.

In order to solve the problems of few available satellites, poor data quality and insufficient monitoring precision in the complex shielding environment in a sluice deformation monitoring process, the invention provides a Beidou / GNSS sluice deformation monitoring data processing method oriented to the complex shielding environment. According to the characteristics that deformation monitoring of a water gate in a complex shielding environment is seriously influenced by a multi-path effect and slow displacement is generated due to the influence of upstream and downstream water levels and seasonal temperature, a dynamic time period coordinate sequence resolving result is obtained by adopting a carrier-to-noise ratio constrained random model, slow displacement constraint observation value robust and half-celestial sphere model multi-path correction method; and carrying out time period displacement extraction based on smooth prior analysis on a dynamic time sequence result, and carrying out fine processing to obtain a time period solution for carrying out real-time and continuous long-term monitoring on deformation and displacement of the sluice. According to the water gate deformation monitoring method in the complex environment, high-precision water gate displacement monitoring with the precision of about 1 mm level can be finally achieved.

The invention relates to the technical field of ocean engineering, and provides a rapid random model correction method for a step-by-step ocean platform structure. According to the design parameters of the ocean platform, establishing an ocean platform initial finite element model; selecting to-be-corrected parameters from the design parameters, and determining the maximum value and the minimum value of each to-be-corrected parameter; performing # imgabs0 # sampling on the maximum value and the minimum value of each to-be-corrected parameter to obtain a corrected parameter matrix; substituting the correction parameter matrix into the initial finite element model to obtain an updated finite element model, obtaining a structure frequency vector corresponding to a correction parameter vector, and establishing a Kriging model; calculating an initial finite element model frequency based on the corrected parameter vector and a Kriging model; constructing a distance objective function and a distribution objective function, and correcting the mean value and the standard deviation of the correction parameters twice; and obtaining a corrected ocean platform finite element model based on the finite element model frequency corrected for the second time.

The invention provides a Beidou PPP-B2b positioning compensation method and system, which are used in the field of satellite navigation, and the method comprises the steps: receiving a broadcast ephemeris, correcting the broadcast ephemeris according to a Beidou satellite PPP-B2b signal to obtain a PPP-B2b precise ephemeris, and obtaining a reference precise ephemeris; according to the PPP-B2b precise ephemeris and the reference precise ephemeris, a PPP-B2b satellite orbit error and a satellite clock error are obtained through calculation, and according to the PPP-B2b satellite orbit error and the satellite clock error, the Beidou satellite PPP-B2b global average SISRE is calculated; respectively calculating a difference SISRE sequence of each satellite epoch based on the global average SISRE of the PPP-B2b, and constructing an SISRE random model based on the difference SISRE sequence of each satellite epoch; the relation between PPP-B2b positioning precision and SISRE random model parameters is measured through experiments, and SISRE optimal process noise parameters are selected from an SISRE random model to compensate a PPP positioning model. According to the scheme, the convergence speed of PPP-B2b positioning can be increased.

The invention is suitable for the technical field of geodetic survey, and provides an InSAR turbulent atmosphere correction method based on a three-dimensional random model, and the method comprises the steps: obtaining M interferograms corresponding to N-scene SAR images of a research region, dividing each interferogram into Y windows through a window with a preset size, taking the phase information in the yth window of all interferograms as the yth observation window; for each observation window, constructing a space-time three-dimensional random model in the observation window according to the M interferograms, constructing a function model of all pixels in the observation window, and solving by using a weighted least square method to obtain a deformation parameter corresponding to the observation window; the deformation parameters corresponding to all the observation windows are spliced, and the deformation phase of the research area is obtained through calculation based on the splicing result. The influence of turbulent atmosphere can be effectively inhibited.

The invention discloses a lithology identification method and system based on TLS, and belongs to the technical field of lithology intelligent identification, and the method comprises the steps: S1, collecting all rock data of a target region, and carrying out the preprocessing of the rock data, and obtaining outcrop point cloud data; s2, constructing a lithology identification unit, and dividing the outcrop point cloud data into point cloud blocks of regular grids according to a preset grid; s3, extracting features with lithologic sensitivity from the point cloud blocks, wherein the features comprise geometric features and spectral features; and S4, inputting the geometric features and the spectral features into a trained random model forest model, and fitting the geometric features and the spectral features by adopting a random forest to obtain lithology categories of various rocks in the target area. According to the method, automatic classification of outcrop lithology is realized, powerful support is provided for geological exploration and lithology analysis, understanding of the geological field is introduced into effective exploration of the artificial intelligence field, and interdisciplinary fusion development is promoted.

The invention discloses the field of automatic control systems, and particularly relates to a random model predictive control method and equipment triggered by a self-adaptive event. The method comprises the following steps: S1, dynamically adjusting the sampling time of a control system and the updating frequency of control input through a self-adaptive event triggering mechanism; the self-adaptive event triggering mechanism is adjusted according to a system state and a change condition of control input; s2, using a distributed robust optimization method to convert a prediction control problem of the random model with opportunity constraint into a second-order conic convex optimization problem for solving; and S3, at each sampling time, determining the optimal control input by solving the second-order cone convex optimization problem, performing feedback control on the state of the control system, and adjusting the communication sampling frequency. According to the method, unnecessary sampling and communication costs are reduced, the feasibility and robustness of the optimization problem are improved, the calculation burden of the system is reduced, and meanwhile, the stability of the control system is ensured.

The invention provides a high-precision precise point positioning method based on BDS-3, and relates to the technical field of wireless communication. The method comprises the following steps: firstly, introducing a traditional ionosphere-free combination model to eliminate the influence of ionosphere delay on positioning precision, and eliminating or weakening a satellite orbit error and a clock error error by using satellite orbit and clock error data provided by an IGS; secondly, constructing a troposphere correction model of adaptive weighted wet delay for eliminating troposphere delay errors; thirdly, establishing a self-adaptive random model, and reasonably weighting pseudo-range and carrier phase observed quantity; and finally, estimating the parameters to be estimated through extended Kalman filtering so as to obtain a positioning calculation result. The method has obvious advantages in the aspect of improving the positioning precision, can meet the requirement of high-precision positioning, can realize centimeter-level positioning precision no matter in a dynamic mode or a static mode, and can even reach millimeter-level positioning precision in the static mode in some observation stations and directions.

The invention relates to a scheduling strategy determination method and device for participation of a wind storage combined system in an electric power spot market, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a day-ahead market predicted electricity price, an intra-day market predicted electricity price and wind power output predicted power in a preset time period, determining an uncertainty set of the day-ahead market electricity price based on the day-ahead market predicted electricity price, obtaining a first sample set of wind power output according to the wind power output predicted power, and obtaining a second sample set of wind power output according to the intra-day market predicted electricity price. Obtaining a second sample set of the daily market predicted electricity price; according to the uncertainty set of the day-ahead market electricity price, the first sample set, the second sample set and a pre-constructed robust random model, determining a power dispatching strategy of the wind storage combined system; and based on the power dispatching strategy, dispatching the wind-storage combined system. By adopting the method, the reliability of a power dispatching strategy can be improved.

The invention relates to a river channel sand body random model generation method and device based on hidden Markov model data filling, a medium and equipment. The method comprises the steps that logging data of a target area are collected and arranged; dividing logging data according to a logging gas-bearing interpretation conclusion, and carrying out analytical statistics on the velocity and density of longitudinal and transverse waves in the logging data; initializing an initial state probability, a state transition matrix, and a mean value and a covariance of Gaussian distribution in the hidden Markov model; inputting a three-dimensional observation vector composed of the velocity and density of the longitudinal and transverse waves into the initialized hidden Markov model for iterative training; performing down-sampling on the lithology data, then constructing a state transition probability matrix, and further constructing a lithology sequence conforming to a geological law through simulation; according to the lithologic sequence, calling a hidden Markov model to generate longitudinal wave velocity, transverse wave velocity and density corresponding to the lithologic sequence; and calculating a reflection coefficient of the synthetic stratum, setting a Ricker wavelet dominant frequency, and obtaining synthetic seismic response data through convolution.

The invention discloses an ocean high-precision time service method and system based on a Beidou satellite, and belongs to the technical field of satellite time service, and the method comprises the steps: building a position parameter random model based on an ocean motion state; establishing a signal-to-noise ratio random model based on the Beidou No.3 PPP-B2b precision and the signal-to-noise ratio of a satellite on the sea; adding the position parameter random model and the signal-to-noise ratio random model into a PPP equation to generate receiver clock difference parameters; and the receiver clock difference parameter is transmitted to the clock calibrator module for clock taming. According to the ocean high-precision time service method and system based on the Beidou satellite, by receiving orbit and clock error correction broadcasted by Beidou third satellite precise point positioning service (PP-B2b), the problems of no Internet, no base station, complex environment, low time service precision and the like in the existing ocean are solved, high-precision single-station time service in a deep and far sea environment is realized, and the time service efficiency is improved. And the time service precision can even reach a subnanosecond level, so that high-precision time information service is provided for establishment of a marine space-time reference.

A multi-modal uncertainty perception mechanical life prediction method based on random model updating comprises the steps that firstly, a drift network is constructed, a representation extraction block RDB comprises four layers including a data structure DS, short-term local feature refinement, long-term time trend development and down-sampling, and residual connection is used for capturing cross-layer information and promoting gradient flow; a diffusion network is constructed, Lipschitz continuity is followed, and ReLU is selected as an activation function; then each subnet is trained, the first two parts of a target function of the subnet are regression terms of a distributed drift network and an adjustment diffusion network, and the latter component belongs to an OOD sample; then, the uncertainty of each subnet is quantified, and finally, multi-modal fusion is carried out based on the respective uncertainty; according to the method, modeling is carried out on the uncertainty of each mode through the fractional order stochastic differential equation sub-network, dynamic fusion is carried out according to the uncertainty of modal characteristics, and a final prediction result and the corresponding uncertainty are obtained through regression reasoning.