Optimization of Random Feature Method in the High-Precision Regime

Machine learning has been widely used for solving partial differential equations(PDEs)in recent years,among which the random feature method(RFM)exhibits spectral accuracy and can compete with traditional solvers in terms of both accuracy and efficiency.Potentially,the optimization problem in the RFM is more difficult to solve than those that arise in traditional methods.Unlike the broader machine-learning research,which frequently targets tasks within the low-precision regime,our study focuses on the high-precision regime crucial for solving PDEs.In this work,we study this problem from the following aspects:(i)we analyze the coeffcient matrix that arises in the RFM by studying the distribution of singular values;(ii)we investigate whether the continuous training causes the overfitting issue;(ii)we test direct and iterative methods as well as randomized methods for solving the optimization problem.Based on these results,we find that direct methods are superior to other methods if memory is not an issue,while iterative methods typically have low accuracy and can be improved by preconditioning to some extent.

Numerical simulation of solitary and random wave propagation through vegetation based on VOF method

A vertical two-dimensional numerical model has been applied to solving the Reynolds Averaged Navier- Stokes （RANS} equations in the simulation of current and wave propagation through vegetated and non- vegetated waters. The k-e model is used for turbulence closure of RANS equations. The effect of vegeta- tion is simulated by adding the drag force of vegetation in the flow momentum equations and turbulence model. To solve the modified N-S equations, the finite difference method is used with the staggered grid system to solver equations. The Youngs＇ fractional volume of fluid （VOF） is applied tracking the free sur- face with second-order accuracy. The model has been tested by simulating dam break wave, pure current with vegetation, solitary wave runup on vegetated and non-vegetated channel, regular and random waves over a vegetated field. The model reasonably well reproduces these experimental observations, the model- ing approach presented herein should be useful in simulating nearshore processes in coastal domains with vegetation effects.

Usefulness of the clip-flap method of endoscopic submucosal dissection: A randomized controlled trial

AIM To prospectively investigate the efficacy and safety of clipflap assisted endoscopic submucosal dissection(ESD) for gastric tumors.METHODS From May 2015 to October 2016, we enrolled 104 patients with gastric cancer or adenoma scheduled for ESD at Shiga University of Medical Science Hospital. We randomized patients into two subgroups using the minimization method based on location of the tumor(upper, middle or lower third of the stomach), tumor size(< 20 mm or > 20 mm) and ulcer status: ESD using an endoclip(the clip-flap group) and ESD without an endoclip(the conventional group). Therapeutic efficacy(procedure time) and safety(complication: Gastrointestinal bleeding and perforation) were assessed. RESULTS En bloc resection was performed in all patients. Four patients had delayed bleeding(3.8%) and two had perforation(1.9%). No significant differences in en bloc resection rate(conventional group: 100%, clip flap group: 100%), curative endoscopic resection rate(conventional group: 90.9%, clip flap group: 89.8%, P = 0.85), procedure time(conventional group: 70.8 ± 46.2 min, clip flap group: 74.7 ± 53.3 min, P = 0.69), area of resected specimen(conventional group: 884.6 ± 792.1 mm^2, clip flap group: 1006.4 ± 1004.8 mm^2, P = 0.49), delayed bleeding rate(conventional group: 5.5%, clip flap group: 2.0%, P = 0.49), or perforation rate(conventional group: 1.8%, clip flap group: 2.0%, P = 0.93) were found between the two groups. Lessexperienced endoscopists did not show any differences in procedure time between the two groups.CONCLUSION For patients with early-stage gastric tumors, the clipflap method has no advantage in efficacy or safety compared with the conventional method.

Novel method for random vibration analysis of single-degree-of-freedom vibroimpact systems with bilateral barriers

The vibroimpact systems with bilateral barriers are often encountered in practice.However,the dynamics of the vibroimpact system with bilateral barriers is full of challenges.Few closed-form solutions were obtained.In this paper,we propose a novel method for random vibration analysis of single-degree-of-freedom(SDOF)vibroim-pact systems with bilateral barriers under Gaussian white noise excitations.A periodic approximate transformation is employed to convert the equations of the motion to a con-tinuous form.The probabilistic description of the system is subsequently defined through the corresponding Fokker-Planck-Kolmogorov(FPK)equation.The closed-form station-ary probability density function(PDF)of the response is obtained by solving the reduced FPK equation and using the proposed iterative method of weighted residue together with the concepts of the circulatory probability flow and the potential probability flow.Finally,the versatility of the proposed approach is demonstrated by its application to two typical examples.Note that the solution obtained by using the proposed method can be used as the benchmark to examine the accuracy of approximate solutions obtained by other methods.

DYNAMIC CHARACTERISTIC ANALYSIS OF FUZZY-STOCHASTIC TRUSS STRUCTURES BASED ON FUZZY FACTOR METHOD AND RANDOM FACTOR METHOD

A new fuzzy stochastic finite element method based on the fuzzy factor method and random factor method is given and the analysis of structural dynamic characteristic for fuzzy stochastic truss structures is presented. Considering the fuzzy randomness of the structural physical parameters and geometric dimensions simultaneously, the structural stiffness and mass matrices axe constructed based on the fuzzy factor method and random factor method; from the Rayleigh＇s quotient of structural vibration, the structural fuzzy random dynamic characteristic is obtained by means of the interval arithmetic; the fuzzy numeric characteristics of dynamic characteristic axe then derived by using the random variable＇s moment function method and algebra synthesis method. Two examples axe used to illustrate the validity and rationality of the method given. The advantage of this method is that the effect of the fuzzy randomness of one of the structural parameters on the fuzzy randomness of the dynamic characteristic can be reflected expediently and objectively.

Numerical investigations on mechanical characteristics and failure mechanism of outwash deposits based on random meso-structures using discrete element method

Outwash deposit is a unique type of geological materials, and its features such as heterogeneity, discontinuity and nonlinearity determine the complexity of mechanical characteristics and failure mechanism. In this work, random meso-structure of outwash deposits was constructed by the technique of computer random simulation based on characteristics of its meso-structure in the statistical sense and some simplifications, and a series of large direct shear tests on numerical samples of outwash deposits with stone contents of 15%, 30%, 45% and 60% were conducted using the discrete element method to further investigate its mechanical characteristics and failure mechanism under external load. The results show that the deformation characteristics and shear strength of outwash deposits are to some extent improved with the increase of stone content, and the shear stress–shear displacement curves of outwash deposits show great differences at the post-peak stage due to the random spatial distribution and content of stones. From the mesoscopic view, normal directions of contacts between "soil" and "stone" particles undergo apparent deflection as the shear displacement continues during the shearing process, accompanying redistribution of the magnitude of contact forces during the shearing process. For outwash deposits, the shear zone formed after shear failure is an irregular stripe due to the movements of stones near the shear zone, and it expands gradually with the increase of stone content. In addition, there is an approximately linear relation between the mean increment of internal friction angle and the stone content lying between 30% and 60%, and a concave nonlinear relation between the mean increment of cohesion and stone content, which are in good agreement with the existing research results.

Modeling habitat suitability of range plant species using random forest method in arid mountainous rangelands

Mountainous rangelands play a pivotal role in providing forage resources for livestock, particularly in summer, and maintaining ecological balance. This study aimed to identify environmental variables affecting range plant species distribution, ecological analysis of the relationship between these variables and the distribution of plants, and to model and map the plant habitats suitability by the Random Forest Method(RFM) in rangelands of the Taftan Mountain, Sistan and Baluchestan Province, southeastern Iran. In order to determine the environmental variables and estimate the potential distribution of plant species, the presence points of plants were recorded by using systematic random sampling method(90 points of presence) and soils were sampled in 5 habitats by random method in 0–30 and 30–60 cm depths. The layers of environmental variables were prepared using the Kriging interpolation method and Geographic Information System facilities. The distribution of the plant habitats was finally modelled and mapped by the RFM. Continuous maps of the habitat suitability were converted to binary maps using Youden Index(?) in order to evaluate the accuracy of the RFM in estimation of the distribution of species potentialhabitat. Based on the values of the area under curve(AUC) statistics, accuracy of predictive models of all habitats was in good level. Investigating the agreement between the predicted map, generated by each model, and actual maps, generated from fieldmeasured data, of the plant habitats, was at a high level for all habitats, except for Amygdalus scoparia habitat. This study concluded that the RFM is a robust model to analyze the relationships between the distribution of plant species and environmental variables as well as to prepare potential distribution maps of plant habitats that are of higher priority for conservation on the local scale in arid mountainous rangelands.

Infinitesimal dividing modeling method for dual suppliers inventory model with random lead times

As one of the basic inventory cost models, the （Q, τ）inventory cost model of dual suppliers with random procurement lead time is mostly formulated by using the concepts of ＂effective lead time＂ and ＂lead time demand＂, which may lead to an imprecise inventory cost. Through the real-time statistic of the inventory quantities, this paper considers the precise （Q, τ） inventory cost model of dual supplier procurement by using an infinitesimal dividing method. The traditional modeling method of the inventory cost for dual supplier procurement includes complex procedures. To reduce the complexity effectively, the presented method investigates the statistics properties in real-time of the inventory quantities with the application of the infinitesimal dividing method. It is proved that the optimal holding and shortage costs of dual supplier procurement are less than those of single supplier procurement respectively. With the assumption that both suppliers have the same distribution of lead times, the convexity of the cost function per unit time is proved. So the optimal solution can be easily obtained by applying the classical convex optimization methods. The numerical examples are given to verify the main conclusions.

Neumann stochastic finite element method for calculating temperature field of frozen soil based on random field theory

To study the effect of uncertain factors on the temperature field of frozen soil, we propose a method to calculate the spatial average variance from just the point variance based on the local average theory of random fields. We model the heat transfer coefficient and specific heat capacity as spatially random fields instead of traditional random variables. An analysis for calculating the random temperature field of seasonal frozen soil is suggested by the Neumann stochastic finite element method, and here we provide the computational formulae of mathematical expectation, variance and variable coefficient. As shown in the calculation flow chart, the stochastic finite element calculation program for solving the random temperature field, as compiled by Matrix Laboratory （MATLAB） sottware, can directly output the statistical results of the temperature field of frozen soil. An example is presented to demonstrate the random effects from random field parameters, and the feasibility of the proposed approach is proven by compar- ing these results with the results derived when the random parameters are only modeled as random variables. The results show that the Neumann stochastic finite element method can efficiently solve the problem of random temperature fields of frozen soil based on random field theory, and it can reduce the variability of calculation results when the random parameters are modeled as spatial- ly random fields.

Response of train-bridge system under intensive seismic excitation by random vibration method

Earthquake is a kind of sudden and destructive random excitation in nature.It is significant to determine the probability distribution characteristics of the corresponding dynamic indicators to ensure the safety and the stability of structures when the intensive seismic excitation,the intensity of which is larger than 7,acts in train-bridge system.Firstly,the motion equations of a two-dimensional train-bridge system under the vertical random excitation of track irregularity and the vertical seismic acceleration are established,where the train subsystem is composed of 8 mutually independent vehicle elements with 48 degrees of freedom,while the single-span simple supported bridge subsystem is composed of 102D beam elements with 20 degrees of freedom on beam and 2 large mass degrees of freedom at the support.Secondly,Monte Carlo method and pseudo excitation method are adopted to analyze the statistical parameters of the system.The power spectrum density of random excitation is used to define a series of non-stationary pseudo excitation in pseudo excitation method and the trigonometric series of random vibration history samples in Monte Carlo method,respectively solved by precise integral method and Newmark-βmethod through the inter-system iterative procedure.Finally,the results are compared with the case under the weak seismic excitation,and show that the samples of vertical acceleration response of bridge and the offload factor of train obeys the normal distribution.In a high probability,the intensive earthquakes pose a greater threat to the safety and stability of bridges and trains than the weak ones.

Reliability analysis for aeroengine turbine disc fatigue life with multiple random variables based on distributed collaborative response surface method

The fatigue life of aeroengine turbine disc presents great dispersion due to the randomness of the basic variables,such as applied load,working temperature,geometrical dimensions and material properties.In order to ameliorate reliability analysis efficiency without loss of reliability,the distributed collaborative response surface method(DCRSM) was proposed,and its basic theories were established in this work.Considering the failure dependency among the failure modes,the distributed response surface was constructed to establish the relationship between the failure mode and the relevant random variables.Then,the failure modes were considered as the random variables of system response to obtain the distributed collaborative response surface model based on structure failure criterion.Finally,the given turbine disc structure was employed to illustrate the feasibility and validity of the presented method.Through the comparison of DCRSM,Monte Carlo method(MCM) and the traditional response surface method(RSM),the results show that the computational precision for DCRSM is more consistent with MCM than RSM,while DCRSM needs far less computing time than MCM and RSM under the same simulation conditions.Thus,DCRSM is demonstrated to be a feasible and valid approach for improving the computational efficiency of reliability analysis for aeroengine turbine disc fatigue life with multiple random variables,and has great potential value for the complicated mechanical structure with multi-component and multi-failure mode.

A Semi-Analytical Method for the PDFs of A Ship Rolling in Random Oblique Waves

The PDFs(probability density functions) and probability of a ship rolling under the random parametric and forced excitations were studied by a semi-analytical method. The rolling motion equation of the ship in random oblique waves was established. The righting arm obtained by the numerical simulation was approximately fitted by an analytical function. The irregular waves were decomposed into two Gauss stationary random processes, and the CARMA(2, 1) model was used to fit the spectral density function of parametric and forced excitations. The stochastic energy envelope averaging method was used to solve the PDFs and the probability. The validity of the semi-analytical method was verified by the Monte Carlo method. The C11 ship was taken as an example, and the influences of the system parameters on the PDFs and probability were analyzed. The results show that the probability of ship rolling is affected by the characteristic wave height, wave length, and the heading angle. In order to provide proper advice for the ship’s manoeuvring, the parametric excitations should be considered appropriately when the ship navigates in the oblique seas.

The improved artificial bee colony algorithm for mixed additive and multiplicative random error model and the bootstrap method for its precision estimation

To solve the complex weight matrix derivative problem when using the weighted least squares method to estimate the parameters of the mixed additive and multiplicative random error model(MAM error model),we use an improved artificial bee colony algorithm without derivative and the bootstrap method to estimate the parameters and evaluate the accuracy of MAM error model.The improved artificial bee colony algorithm can update individuals in multiple dimensions and improve the cooperation ability between individuals by constructing a new search equation based on the idea of quasi-affine transformation.The experimental results show that based on the weighted least squares criterion,the algorithm can get the results consistent with the weighted least squares method without multiple formula derivation.The parameter estimation and accuracy evaluation method based on the bootstrap method can get better parameter estimation and more reasonable accuracy information than existing methods,which provides a new idea for the theory of parameter estimation and accuracy evaluation of the MAM error model.

Accelerated Matrix Recovery via Random Projection Based on Inexact Augmented Lagrange Multiplier Method

In this paper, a unified matrix recovery model was proposed for diverse corrupted matrices. Resulting from the separable structure of the proposed model, the convex optimization problem can be solved efficiently by adopting an inexact augmented Lagrange multiplier (IALM) method. Additionally, a random projection accelerated technique (IALM+RP) was adopted to improve the success rate. From the preliminary numerical comparisons, it was indicated that for the standard robust principal component analysis (PCA) problem, IALM+RP was at least two to six times faster than IALM with an insignificant reduction in accuracy; and for the outlier pursuit (OP) problem, IALM+RP was at least 6.9 times faster, even up to 8.3 times faster when the size of matrix was 2 000×2 000.

RANDOM MICROSTRUCTURE FINITE ELEMENT METHOD AND ITS VERIFICATION FOR EFFECTIVE PROPERTIES OF COMPOSITE MATERIALS

The present study aims at developing a new method-RandomMicrostructure Finite Element Method (RMFEM) for the effectiveproperties of com- Posite materials. In this method, a randommicrostructure Model is used to simulate the microstructure of thereal composite materials. The physical fields in such a randomMicrosturucture model under specified boundary and initial Conditionsare analyzed by finite element method.

Comparison of finite difference and pseudo-spectral methods in forward modelling based on metal ore model of random media

With more applications of seismic exploration in metal ore exploration,forward modelling of seismic wave has become more important in metal ore. Finite difference method and pseudo-spectral method are two important methods of wave-field simulation. Results of previous studies show that both methods have distinct advantages and disadvantages: Finite difference method has high precision but its dispersion is serious; pseudospectral method considers both computational efficiency and precision but has less precision than finite-difference. The authors consider the complex structural characteristics of the metal ore,furthermore add random media in order to simulate the complex effects produced by metal ore for wave field. First,the study introduced the theories of random media and two forward modelling methods. Second,it compared the simulation results of two methods on fault model. Then the authors established a complex metal ore model,added random media and compared computational efficiency and precision. As a result,it is found that finite difference method is better than pseudo-spectral method in precision and boundary treatment,but the computational efficiency of pseudospectral method is slightly higher than the finite difference method.

Random failure mechanism method for assessment of working platform bearing capacity with a linear trend in undrained shear strength

A bearing capacity evaluation for the surface strip foundation on a working platform modelled on a twolayered substrate is considered in the study.The upper layer is assumed as man-made and wellcontrolled and thus non-variable.The lower layer modelling natural cohesive soil is subjected to spatial variability of undrained shear strength.The random failure mechanism method(RFMM)is used to evaluate the bearing capacity.This approach employs a kinematic assessment of the critical load and incorporates the averaging of three-dimensional(3 D)random field along dissipation surfaces that result from the failure mechanism geometry.A novel version of the approach considering an additional linear trend of undrained shear strength in the spatially variable layer is proposed.The high efficiency of the RFMM algorithm is preserved.The influences of foundation length,trend slope in the spatially variable layer,fluctuation scales,and thickness of the homogenous sand layer on the resulting bearing capacity evaluations are analysed.Moreover,for selected cases,verification of the RFMM based assessment obtained using random finite difference method(RFDM)based on 3 D analysis is provided.Two types of analyses are performed using RFDM based on associated and non-associated flow rules.For associated flow rule which corresponds to RFMM,the RFMM is conservative and efficient and thus it seems preferable.However,if RFDM employs non-associated flow rule(much lower dilation angle for sand layer),the efficient RFMM is no longer conservative.For this situation,a combined approach that improves the efficiency of the numerical method is suggested.

An Improved Non-Parametric Method for Multiple Moving Objects Detection in the Markov Random Field

Detecting moving objects in the stationary background is an important problem in visual surveillance systems.However,the traditional background subtraction method fails when the background is not completely stationary and involves certain dynamic changes.In this paper,according to the basic steps of the background subtraction method,a novel non-parametric moving object detection method is proposed based on an improved ant colony algorithm by using the Markov random field.Concretely,the contributions are as follows:1)A new nonparametric strategy is utilized to model the background,based on an improved kernel density estimation;this approach uses an adaptive bandwidth,and the fused features combine the colours,gradients and positions.2)A Markov random field method based on this adaptive background model via the constraint of the spatial context is proposed to extract objects.3)The posterior function is maximized efficiently by using an improved ant colony system algorithm.Extensive experiments show that the proposed method demonstrates a better performance than many existing state-of-the-art methods.

STUDY ON METHOD OF NUMERICAL SIMULATION OF NONLINEAR RANDOM WAVES

This new method can simulate the nonlinear random wavcs processes by computer if the higherorder moments of the probability distribution of the sea surface elevation reflecting the nonlinearity ofthe sea wave are given. Compared with other methods, this method has greater accuracy andflexibility, wider application and faster simulation. Statistical analysis of the sea surface elevationdistribution of the simulated wave process showed obviously the Gram-Charlier series can be used to depictthe distribution of the sea surface elevation.

An Application of Direct Ritz Method for Random Seismic Responses of Jacket Platforms

In this paper, the jacket platform is simulated by a non-uniform cantilever beam subjected to axial force. Based on the Hamilton theory, the equation of bending motion is developed and solved by the classical Ritz method combined with the pseudo-excitation method for random responses with non-classical damping. Usually, random responses of this continuous structure are obtained by orthogonality of modes, and some normal modes of the structure are needed, causing inconvenience for the analysis of the non-uniform beam whose normal modes are not easy to be obtained. However, if the pseudo-excitation method is extended to calculate random responses by combining it with the classical Ritz method, the responses of a non-uniform beam, such as auto-PSD function, cross-PSD and higher spectral moments, can be solved directly avoiding the calculation of normal modes. The numerical results show that the present method is effective and useful in aseismic design of platforms.