44 results about "Model decomposition" patented technology

The invention relates to an image restoring method, which comprises the following steps: firstly carrying out direction and experience model decomposition on a target image, allowing to draw damaged edge structure information in a given area to be restored according to the existing information, utilizing a frequency characteristic value obtained by the direction and experience model decomposition to calculate similarity so as to firstly finish edge structure restoration in the whole damaged area, then combining degree of confidence to calculate the priority of a residual filling area according to gradient characteristics of pixel points on the boundary of the area to be restored, and utilizing a frequency characteristic value obtained by the decomposition again to calculate similarity so as to finish the restoration of corresponding areas. The method not only can restore the structure and texture information in a larger-scale defective area, but also can restore the images with certain sheltering relation. The robustness in the restoring process is high, thus effectively preventing continuous expansion of error information.

The invention relates to an oil/gas/water multi-phase flow type identification method based on main constituent analysis and supporting vector machine. The feature is that: taking collection to flow parameter signal of different flow type by using horizontal tubes, pressure sampling ring, pressure sampling tube, pressure difference transmitter, data collection card and computer; taking denoising processing to pressure difference fluctuation signal between the collected different pressure sampling distance; taking empiric model decomposition to the signal to gain the fixed modal function, and forming a matrix X from the modal functions, using the main constituent analysis to gain the eigenvector of the flow type; using the supporting vector machine to accomplish the mapping from feature space to flow type space to realize the final flow type identification. The eigenvector fusing the information from plural sensors and would fully reflect the information of flow type. It has the feature of rapid learning speed and strong classifying ability.

The invention provides a 3dsMax-based nuclear facility model radiation field dosage simulation method. The method comprises the following steps of: constructing a model by 3dsMax software according to determined nuclear facility parameters, and storing a file in a 3DS format; importing a 3DS nuclear facility model file and obtaining model parameters; decomposing the 3dsMax nuclear facility model into voxels by using an octree method; writing determined voxel parameters and material information into an input card; importing the input card into a point nuclear integration program; calculating an accumulation factor; calculating a mean free path, in a radiation field, of a gamma ray; establishing a flux rate-dosage rate conversion factor, quality decay factors of chemical elements and materials and a single-layer accumulation factor database by utilizing an SQLite database engine; carrying out combined operation on a box by using a Boolean connective operator, and constructing a complicated radiation field geometric structure; and calculating a three-dimensional radiation field dosage by using a point nuclear integration method. The method provided by the invention is capable of realizing radiation field dosage calculation of complicated 3dsMax nuclear facility models with sizes, materials and energy parameters.

The invention relates to a recommendation prediction method based on attribute information reference self-leaning. Attribution information is utilized for slowing down cold start, the advantages of user and commodity collaborative filtering and model decomposition based on a matrix are combined, the training speed is high, and the interpretability is achieved; meanwhile, on the condition that rating data is sparse, the prediction precision is superior to that of user and commodity collaborative filtering and the matrix-based decomposition method.

The invention discloses a region saturation load prediction method based on a model family decomposition and integration technology. In the existing load prediction technology, single prediction methods are used, such as trend extrapolation, grey prediction and linear regression; and the prediction results for each prediction method are different greatly. The region saturation load prediction method includes the steps: extracting the main influence factors for region load growth; selecting a suitable single prediction method or model to predict region saturation load; and finally determining the weight of each single prediction method or model so as to perform combined prediction of the region saturation load scale. The region saturation load prediction method based on a model family decomposition and integration technology integrates with the advantages of each single saturation load prediction method, can perform optimized combination and comprehensive analysis, can comprehensively consider the influence of the indexes which are closely related with power consumption, integrates with the saturation load determination index set to predict the saturation scale and the saturation year of the region electric power consumption quantity, and is conductive to assisting long term of power grid planning in the region.

The invention discloses a multicore parallel solving method for the computation of polymer molecular weight distribution, comprising the following steps of: establishing a mechanism model of free radical polymerization dynamic simulation to solve the polymer molecular weight distribution by taking the free radical polymerization dynamic simulation as an object; dividing the mechanism model into a small-scale free radical polymerization model and a large-scale free radical polymerization model through a decoupling method; solving the small-scale free radical polymerization model by utilizing a variable step-size and order backward difference method; and solving the large-scale free radical polymerization model to accelerate the solution of the computation of the polymer molecular weight distribution through a parallel sequential method by taking a multicore computer system as a computing platform. Compared with the traditional non-parallel solving method of the computation of the polymer molecular weight distribution, the invention can increase the solving speed of the mechanism model by sufficiently utilizing the multicore characteristics of the multicore computer system; and in addition, the multicore parallel solving method has simple and clear principle and is convenient to implement on any multicore computer system.

The invention provides a complete polarization synthetic aperture radar target decomposition method for adaptive selection unitary transformation. The method comprises the following steps: (1) performing two unitary transformations for singh for data coherence T matrix of the complete polarization synthetic aperture radar to obtain the matrix, (the formula is as shown in specification); (2) performing other two unitary transformations for the coherence T matrix to obtain the matrix T(omega), wherein the first unitary transformation is used for performing the spiral angle compensation and restraining the volume scattering excessive estimation of model decomposition, the second unitary transformation is used for further restraining the volume scattering excessive estimation and reducing one degree of freedom of the coherence T matrix; (3) comparing with element (the formula is as shown in specification) of the matrix (the formula is as shown in specification) with the element T33(omega) of the matrix T(omega); if (the formula is as shown in specification), and (the formula is as shown in specification), otherwise, T is equal to T(omega); (4) performing three-component model decomposition on coherence T matrix. The two unitary transformations for singh in the step (1) or the two unitary transformations in the step (2) can be selected for the coherence T matrix in a self-adaption mode by the method according to the real situation of the object, and the volume scattering excessive estimation problem of the model decomposition can be effectively restrained.

The invention discloses an electric power system optimal support set positioning method based on anticipated fault decomposition. In order to stabilize the frequency of a power grid, multiple units having decisive effects on safety and reliability of the power grid need to be selected. The method comprises steps of by considering different anticipated faults of the power grid, establishing an electric power system safety restraining optimal power flow mode by taking the minimization of the climbing ability of generator sets as an objective; adopting an anticipated fault decomposition method to decompose the model into a main problem under a normal operation mode and sub problems under each anticipated fault operation mode; through iteration solving of the main problem under the normal operation mode and the sub problem under each anticipated fault operation mode, achieving solving of above electric power system safety restraining optimal power flow mode; and according to the steps, solving the smallest climbing ability transformation quantity of the generator sets, carrying out progressively decreased ranking on the generator sets with the non-zero transformation quantity and successively recording the generator sets to be the optimal support sets with progressively decreased priorities according to the sequences to be output as the results. The method is excellent in applicability and actual requirements are well met.

The invention discloses a polarization SAR model decomposition evaluation method based on electromagnetic scattering simulation. The method comprises the steps of (S1) setting surface electromagneticgeometric parameters, aircraft flight parameters and radar satellite position parameters and calculating a surface real beta value of a target scene by using a classical forward model, (S2) processingthe polarization SAR simulation data of the target scene by using a polarization SAR model decomposition method to be evaluated and carrying out inversion to obtain an inversion beta value, and (S3)calculating a mean square root error of the inversion beta value and the surface real beta value, and evaluating the polarization SAR model decomposition method to be evaluated with a principle that the smaller the mean square root error is, the better the effect of the decomposition method is. According to the scheme of the invention, the electromagnetic scattering simulation theory and the polarization SAR model decomposition theory are organically combined, a decomposition algorithm is evaluated from the perspective of electromagnetic wave simulation and model decomposition, the method is fair and just, and a reference can be provided for the selection of an excellent model decomposition method for different application scenes.

The invention puts forwards a 3D (Three-dimensional) printing-oriented model decomposition and arrangement method. The method comprises the following steps: giving a model S and a printing space PV, calculating the decomposition of the initial pyramid attribute blocks of the model S, carrying out voxelization on the pyramid attribute blocks to generate a first initial intermediate solution isolu0; initializing an intermediate solution set ISolu={isolu0}, and initializing an optimal complete solution P*; for each intermediate solution isolui in the intermediate solution set ISolu, calculating one series of candidate intermediate solutions generated by the intermediate solutions, and solving an optimal decomposition and arrangement solution of the model S; and carrying out local optimization on a three-dimensional grid of a plurality of extracted blocks, causing the blocks to be more tightly arranged through the translation of each block along the negative direction of a Z axis, and therefore, obtaining the optimal decomposition and arrangement about the model S. The model is decomposed into few blocks, the few blocks are effectively arranged into a printing space, and therefore, a printing process is effectively carried out.

The invention discloses a gap multi-model weighting function parameter self-tuning method, which is characterized in that the characteristics of a nonlinear system are analyzed, scheduling variables capable of reflecting system operating conditions are selected, a model set of an approximate nonlinear system is constructed by using a multi-model decomposition algorithm, a sub-controller is designed based on each sub-model in the model set, a weighting function of the sub-controller comprises a reciprocal of a gap distance between the nonlinear system and each sub-model at a current moment, theweighting function has a unique tuning parameter, and an optical value of the tuning parameter is obtained by optimizing an integral absolute error value between closed-loop output and reference input of the nonlinear system. The gap multi-model weighting function parameter self-tuning method has the advantages of avoiding the cumbersomeness of manual parameter tuning of the weighting function, reducing the dependence of parameter tuning on prior knowledge, and optimizing the control performance of a closed-loop system.

The invention provides a satellite orbit determining method and device based on maximum model decomposition. The satellite orbit determining method comprises the steps that based on the KAM principle, a Fourier series expansion model of position coordinates of a satellite is established when the satellite operates on an orbit; according to the descending order of amplitudes, the Fourier series expansion model of the position coordinates is decomposed, and a decomposed similar series expansion model of the position coordinates is obtained; based on the actual observation values of the position coordinates of the satellite when the satellite operates on the orbit, the amplitudes and the corresponding frequencies in the similar series expansion model of the position coordinates are determined; through the similar series expansion model of the position coordinates, the position coordinates of the satellite on the orbit at any moment are predicated. The satellite orbit is determined more rapidly and conveniently by the adoption of the satellite orbit determining method and device based on maximum model decomposition, the satellite orbit determining method and device based on maximum model decomposition is practical, only the frequencies and the amplitudes of the satellite orbit need to be computed and stored in the predication process, storage resources are saved, and the processing efficiency is greatly improved.

The invention discloses a sea wave height analysis method. The sea wave height analysis method comprises a fast band-pass filtering method for signal mode decomposition; a model decomposition trend is used for substituting for a sectioned polynomial fitting trend in an MF-DFA (multifractal detrended fluctuation analysis) method. A partition function method for multifractal analysis and the MF-DFA method are applied to sea wave height testing data analysis to indicate that a wave height sequence has a weak multi-fractal characteristic. The sea wave height analysis method makes improvements aiming at the problems existing in the MF-DFA method and establishes an MF-DFA method based on signal mode decomposition; testing data verification indicates that the improved method has the capability of avoiding the defects of the original method and has certain advantages on the basis of well meeting an elimination trend of the original method.

The invention relates to a big data sharing-based 3D printing platform and a working method thereof. The 3D printing platform comprises a cloud server, a 3D printer and an upper computer connected with the cloud server, wherein the upper computer is suitable for inputting printing modules and storing decomposed data of to-be-printed models to the cloud server, suitable for obtaining decomposed data of corresponding printing models from the cloud server, and further suitable for carrying out 3D printing on the obtained decomposed data of the printing models through the 3D printer; data decomposition is carried out on the input printing models through the upper computer and the decomposed data is stored in the cloud server, so that the data can be conveniently called by other upper computers; and a big data sharing distribution mechanism is further combined, so that sharing distribution is carried out on the decomposed data of the printing models, and data search and matching are carriedout by using pictures of the printing models.