53 results about "Design matrix" patented technology

The invention discloses a tip timing vibration parameter identification method for a rotating blade. The method comprises the following steps of determining the number of tip timing sensors and circumferential installation positions; collecting an under-sampled signal of a vibration displacement of the rotating blade; building a sparse representation model of the under-sampled signal, and determining a vibration frequency of the blade by using a sparse reconstruction method; and obtaining a vibration order of the blade based on the vibration frequency of the blade determined by the sparse reconstruction method and a rotation frequency of the blade, building a blade vibration equation design matrix based on the vibration order of the blade and the installation positions of the tip timing sensors, and identifying a vibration amplitude of the rotating blade by using a circumferential Fourier method. The invention further provides a tip timing vibration parameter identification system forthe rotating blade. According to the method and the system, no additional prior information is needed; the vibration frequency and amplitude parameters of the blade can be identified by measuring theunder-sampled signal of the vibration displacement only by utilizing the tip timing sensors; the vibration parameter identification step is simplified; and the precision of an identification result ishigh.

The present invention provides an actuarial modeling method based on the car body combination property index of support vector machine, belonging to auto structure optimization design field, including step 1, selecting parts as design object, and determining thickness of parts, material parameter as design variable; step 2, adopting testing design method to generate design matrix according to the design variable determined by step 1; step 3, processing finite element simulation testing design according to the design matrix, extracting structural performance index value to complete the establishment of the training sample; step 4, training sample by supporting vector machine method, establishing combination property index mathematic model; step 5, verifying the established combination property index mathematic model by checking index; step 6, establishing adaptive process after verifying the high precision combination property index mathematic model. The invention is fit for nonlinear response index mathematical modeling having multiple variables, high structural performance, having high precision of the model prediction, high efficiency of modeling, high feasibility of optimum design.

A design matrix structure visualization technique and tool for the study and analysis of systems, such as software systems, where the bounds of the matrix are defined by a selected focusing metric and entities forming the system are displayed in the matrix as blocks of size relative to their proportion of the selected focusing metric. Relationships between entities are illustrated by summing all relationships between individual elements forming each displayed entity. Where possible, an identifier defining the characteristic of a displayed block is also shown. Preferably, the technique and tool includes a “zoom” feature to allow a user to drill down and uncover details associated with smaller blocks as shown in the original view.

A method of identifying differentially-expressed genes includes deriving an analysis of variance (ANOVA) or analysis of covariance (ANCOVA) model for expression data associated with a number of genes; from the ANOVA or ANCOVA model, deriving a linear regression model defined at least in part by an observation vector representative of an observed subset of the gene-expression data, a design matrix of regressor variables, a vector of regression coefficients representing gene contribution to the observation vector, and a measurement error vector; and to the linear regression model, applying a hierarchical selection algorithm to designate a subset of the regression coefficients as significant regression coefficients, the selection algorithm representing at least one of the observation vector, the design matrix, and the measurement error vector as being hierarchically dependent on parameters having predetermined probabilistic properties, wherein the designated subset corresponds to a respective subset of the genes identified as differentially expressed.

The invention provides a GNSS point positioning method and device based on Kalman filtering. The GNSS point positioning method comprises the steps that a dynamic model and an observation model of Kalman filtering are built; state values of first prediction and a variance matrix of the state values of first prediction are obtained according to the built dynamic model and observation model respectively; a gain matrix is calculated according to the obtained variance matrix of the state values of first prediction, a design matrix and an observation error variance matrix; the state values of second prediction are calculated according to the obtained state values of first prediction, the design matrix, observation values and the gain matrix; the variance matrix of the state values of second prediction is calculated according to the obtained variance matrix of the state values of first prediction, the gain matrix and the design matrix; smooth speed is obtained, and the state values of the next moment and the variance matrix of the state values are calculated according to the obtained smooth speed, the state values of second prediction and the variance matrix of the state values of second prediction. According to the technical scheme, precision and usability of a positioning result are improved.

The invention discloses an arrangement method for a leaf apex timing sensor used for distinguishing blade multi-modal vibration. The method comprises the following steps that: measuring the revolvingspeed and the vibration frequency of a blade, on the basis of the revolving speed and the vibration frequency of the blade, determining the vibration order of the blade and the number of the leaf apextiming sensors; on the basis of the installation angle of the leaf apex timing sensor and the vibration order, constructing the design matrix of a blade vibration model; calculating the condition number of the design matrix; and taking the condition number of the design matrix as a target function, and utilizing an optimization algorithm to enable the condition number of the design matrix a to achieve a minimum value so as to determine the installation angle of the leaf apex timing sensor.

The invention discloses a trans-scale design method of a high-speed milling cutter and the milling cutter. At present, the research of the internal relation among a microstructure, a macrostructure and performance of cutter materials is absent, and scientific evidences of the design and the development of the novel high-speed milling cutter are absent. The method comprises the following steps of: (1) constructing a high-speed milling cutter safety decline behavioral characteristic model; (2) developing the high-speed milling cutter by using a high-speed milling cutter safety stability design model; (3) establishing a high-speed milling cutter safety decline behavioral characteristic model design matrix by a grey correlation analysis method to characterize the relation between milling cutter design parameters and safety decline; (4) establishing a high-speed milling cutter mesoscopic level safety model by a grey cluster analysis method; and (5) researching macroscopic and mesoscopic mechanical characteristics of a high-speed milling cutter component under the condition of presetting an external force boundary, and realizing force connection-based high-speed milling cutter trans-scale correlation through continuous medium-molecular dynamic characteristic zigzag mapping. The method is used for designing the high-speed milling cutter.

An optimum measurement subset with a specified number of elements is generated from a set of input global navigation satellite system (GNSS) measurements. A design matrix and a weight matrix are generated. Values of a set of coefficients corresponding to the set of input GNSS measurements are calculated. The value of a specific coefficient is calculated as the ratio of the change in value of the at least one target parameter resulting from the change in value of the specific input GNSS measurement to the change in value of the sum of squared residuals resulting from the change in value of the specific input GNSS measurement. The optimum measurement subset is selected based at least in part on the values of the set of coefficients.

An optimum measurement subset with a specified number of elements is generated from a set of input global navigation satellite system (GNSS) measurements. A design matrix and a weight matrix are generated. Values of a set of coefficients corresponding to the set of input GNSS measurements are calculated. The value of a specific coefficient is calculated as the ratio of the change in value of the at least one target parameter resulting from the change in value of the specific input GNSS measurement to the change in value of the sum of squared residuals resulting from the change in value of the specific input GNSS measurement. The optimum measurement subset is selected based at least in part on the values of the set of coefficients.

The invention discloses a relay service life prediction method based on multidimensional design parameters, and the method comprises the steps: firstly analyzing the working principle of a relay product, and obtaining relay service life experiment data under different design parameter conditions through a variable control method; then analyzing and determining sensitive design parameters influencing the service life of the relay; in the process of carrying out the service life experiment, collecting the working data of the relay, and determining characteristic parameters representing the service life of the relay; establishing a regression design matrix, taking the relay life characteristic parameters as observation vectors, solving parameter vectors, and obtaining a mathematical model ofthe characteristic parameters about the design parameters; taking the change rate of the characteristic parameters as an observation vector, and fitting to obtain a mathematical model of the change rate of the characteristic parameters relative to the design parameters; and finally, combining the two models to obtain a life prediction model. The relation between the relay service life and the multi-dimensional design parameters can be determined through modeling, and relay design can be guided to further prolong the relay service life.

An ambiguity fixing method based on sequential least squares of the present invention can improve the success rate of measuring the ambiguity between two antennas in real time by using the single-frequency or double-frequency navigation satellite signals in a dynamic environment, provides the data support for the subsequent high-precision baseline measurement, attitude determination and the like,and is characterized by introducing the priori information in the ambiguity real number solution estimation process to serve as the state information to be transmitted to a filter, and accelerating convergence while reducing the morbidity of a design matrix. According to the method, the thought based on a measurement error modeling theory and a matrix triangulation decomposition technology is applied under a sequential condition, the additional calculation burden is not increased while the measurement information is accumulated, an ambiguity real number solution and the cofactor matrix information are calculated by constructing an ambiguity filter in a least square form, and finally, the ambiguity real number solution information is processed by using an LAMBDA method to search an ambiguity integer solution, thereby solving the problem of low ambiguity solving success rate of navigation satellite signals in a dynamic environment in a traditional method.

The invention relates to a DSM-based Method for analyzing change of a supply chain system. The method comprises that an initial design matrix A is obtained according to relation among different change parameters in a supply chain network; the initial design matrix A is transformed in a reachable way to obtain a reachable matrix P; clustering analysis is carried out on an initial design matrix M to obtain a clustering matrix C; the change parameters are layered, and change paths are classified; and XML is used to store path information. The graph theory is used to analyze change, relates among changed parameters and parameters are represented by nodes and directed sides in a digraph, DSM is established to carry out clustering analysis and further to predict a possible change path, and the path is stored in the XML structure. The change occurrence point and spreading and amplifying range are mastered generally, and change is controlled effectively according to concrete conditions.

The invention relates to a multi-frequency multi-mode GNSS generalized absolute code bias estimation method. The method includes the steps of redefining and classifying the current multi-frequency multi-mode GNSS code bias, constructing a geometry-free observation equation and an ionosphere-free observation equation respectively by using original pseudo-range observation data of a multi-frequencymulti-mode GNSS observation network, fully expressing the code bias in the observation equations, according to the discriminability of various bias parameters in a design matrix, recombining the biasparameters combined with the condition number of the matrix rank deficiency, and estimating the code bias and other parameters by least squares; and by introducing the same observation standard as anexisting clock difference product, converting a relative code bias parameter into an absolute code bias parameter. The method of the invention can provide a quasi-absolute code bias value for an original observation value, provides a simple and uniform method of bias correction for navigation and positioning users, and greatly reduces the complexity of multi-frequency multi-mode GNSS data processing.

The invention provides a brain functional region positioning method based on local smoothing regressions. The brain functional region positioning method based on the local smoothing regressions comprises the following steps of pretreating data and deciding a design matrix X; taking a voxel vi as the center of a sphere and r as a semidiameter for the establishment of a spherical selected region and extracting the time sequence of all the voxels in the spherical selected region; according to the time sequence of all the voxels in the spherical selected region and the design matrix, forming an objective function and optimizing the objective function; calculating a condition specificity effect of the voxel vi; turning to the next voxel vi+1 and repeating steps from S2 to S4 till the execution of the steps on each voxel of a whole brain; and setting a threshold value for a whole brain perception mapping so as to obtain a brain functional region positioning map relevant with stimulus conditions. All the generalized linear models based on the regressions of the single voxels and based on Gaussian smoothing filtering can be regarded as special cases of the invention; the brain functional region positioning method based on the local smoothing regressions can be integrated into a framework used for a searchlight method; after the obtainment of regression coefficients, mahalanobis distances between various predictor coefficients are calculated; and through the adjustment on hyper-parameters alpha and beta, smoothing effects of various degrees are obtained.

The present invention provides a method for calculating the body temperature of cattle by skin temperature. The method comprises steps of: calculating a rectal temperature of the cattle by using a mixed linear model as shown in model 1 referring to Y=a+X0b0+Xsbs+Xtbt+Xibi+e, wherein Y is the rectal temperature, X0 is the skin temperature, Xs is a season design matrix, bs is a non-random parameter vector of Xs, Xt is a time design matrix, b0 is a regression coefficient of X0, bt is a non-random parameter vector of Xt, Xi is an individual design matrix, bi is a random parameter vector of Xi, a represents the intercept, and e is a random vector error. According to the method, in the regression analysis of the rectal temperature by employing the skin temperature, the seasonal factor and the time point factor are taken as the fixed effects, and the individual factor is taken as the random effect. The skin temperature after correction by employing the model is compared with the actual rectal temperature, the average of the absolute numbers of differences between the correction temperatures and the rectal temperature reaches 0.104 DEG C, and the absolute values of 90% of data differences are not higher than 0.2 DEG C, and the ideal effect is achieved.

A design matrix structure visualization technique and tool for the study and analysis of systems, such as software systems, where the bounds of the matrix are defined by a selected focusing metric and entities forming the system are displayed in the matrix as blocks of size relative to their proportion of the selected focusing metric. Relationships between entities are illustrated by summing all relationships between individual elements forming each displayed entity. Where possible, an identifier defining the characteristic of a displayed block is also shown. Preferably, the technique and tool includes a “zoom” feature to allow a user to drill down and uncover details associated with smaller blocks as shown in the original view.

The invention relates to a RTK (Real-time kinematic) tilt measurement accuracy detection method and system, computer equipment and a computer storage medium. The above RTK tilt measurement accuracy detection method comprises steps: RTK tilt measurement data acquired for RTK tilt measurement for a measurement target point are acquired; a distance resection adjustment model is constructed accordingto the RTK tilt measurement data, and a design matrix, a prior variance matrix and an observation value-calculation value vector are determined, wherein the distance resection adjustment model is a model for characterizing geodetic coordinate series residual data and variance data; and according to the prior variance matrix and the observation value-calculation value vector, the position accuracyof a measurement target point is calculated, a geometric accuracy factor is calculated according to the design matrix, and the measurement target point coordinate estimation accuracy is detected according to the position accuracy and the geometric accuracy factor. On the basis of ensuring high-accuracy RTK tilt measurement data extraction stability, and the high-accuracy measurement target point coordinate estimation cost is reduced.

The invention discloses a method for recovering an LFM signal under short-time strong interference. The method comprises the following steps: I, designing a reference threshold according to a signal amplitude, detecting the threshold of a received signal, and recognizing the starting and stopping time of the strong interference; II, extracting a random unit array column vector by avoiding an interference moment, and finishing compressed sampling while filtering interference from the LFM signal with short-time strong interference by using the designed matrix; III, processing the processed dataaccording to a sparse component norm maximum value method reconstructed by compressed sensing to obtain a FrFT optimal transformation order p; IV, performing p-order FrFT calculation on the column vector of a unit array to obtain a sparse vector base; and V, reconstructing the signal by using a compressed sensing sparse recovery algorithm based on the sparse vector base to obtain a signal frequency modulation rate and starting frequency. The method can be used for recovering a linear frequency modulation signal for a condition background with short-time strong interference.

The invention provides a data collection device and a data processing method. The device comprises a bottom box and a top cover; wherein the bottom box comprises an outer box, a heat preservation layer and an inner box; a bottom plate is arranged in the inner box, and the bottom plate divides the inner box into a lower airflow space and an upper test space; partition plates are respectively arranged at two ends in the test space to divide the test space into two wind power areas and a test area; and mounting openings are formed in the longitudinal direction of the partition plates, wind powercomponents are installed in the mounting openings, and shielding components are arranged on outer sides of the mounting openings. The data collection device provided by the invention can realize the simulation of various environmental parameters; the wind power can be adjusted up and down through the wind power components, on-demand supply of wind power is realized by the shielding components, andthe simulation effect is good. Based on the introduction of a multicollinearity diagnosis method of a global model, a weighted variance inflation factor method and a weighing condition index method-decomposition ratio method are constructed to diagnose the multicollinearity problem of a design matrix of a geographic weighting regression model.

A system, method and computer program product for quickly and approximately solving structured regression problems. In one aspect, the system, method and computer program product are applied to problems that arise naturally in various statistical modeling settings—when the design matrix is a Vandermonde matrix or a sequence of such matrices. Using the Vandermonde matrix structure further accelerates the solution of the regression problem, achieving running times that are faster than “input sparsity”. The modeling framework speedup benefits of randomized regression for solving structured regression problems.

The invention discloses a modularization design matrix type electric connector. The electric connector comprises a standard socket and plugs composed of contact member seats and a plug casing, wherein the plug casing is composed of a support, a cover board and a locking clamping hook, one or more contact member seats are sequentially parallelly arranged on the support to form the plugs with the different cores, the plugs are arranged above the standard socket, and contact members on the plugs are plugged in socket holes of the standard socket in rows and columns. The electric connector is advantaged in that one standard socket arranged in a matrix mode is employed, the standard socket can be plugged with the plugs composed of one contact member seat or multiple contact member seats, the one standard socket to multiple-plug plugging effect is realized, the plugs with the different cores can realize selective joint with the socket, problems of complex product test wiring and non-easy management and maintenance are solved, and production efficiency of electronic product tests is improved.

The invention relates to a resistance model modeling method, wherein the method comprises the following steps: 1) designing a matrix type resistance test piece according to a graphic design rule, andforming a resistance box model by adjacent four resistance sizes; 2) measuring a current and voltage characteristic curve of the matrix resistance test piece; 3) extracting model parameters of each size resistor according to a measurement result; 4) calculating the model parameters of each resistance box model according to the resistance model parameters; 5) synthesizing the model parameters of the resistance box models; and 6) extracting the temperature coefficient based on high and low temperature data. The resistance model modeling method provided by the invention can improve the model accuracy and save the modeling time.

The invention discloses a method for obtaining optimized parameters of a multi-ring sleeved tungsten alloy flywheel. The method comprises the following steps: establishing a parameterized finite element model of the multi-ring sleeved tungsten alloy flywheel by using ABAQUS software; establishing a parameterized finite element analysis file of the multi-ring set tungsten alloy flywheel by using ABAQUS software; adopting iSIGHT optimization design software for building a DOE test design model, and obtaining a test design matrix; calling ABAQUS software to analyze and calculate each data point in the test design matrix by writing. Bat files to generate DOE sample data; based on DOE sample data, adopting an RBF radial basis function to construct a neural network agent model, obtaining a mathematical model between factors and responses, obtaining the interference magnitude optimization value between a tungsten alloy middle layer and an outer retaining ring with the allowable stress value as the optimization target by mining the internal relation of the data under the constraint condition, and obtaining the optimal interference magnitude of the tungsten alloy middle layer and the outerretaining ring. The size optimization values of the inner hub, the tungsten alloy middle layer and the outer retaining ring are optimized by taking the energy storage density and the rotational inertia as optimization targets.

The invention provides an implant intervention medical instrument key parameter extraction and evaluation method based on axiom design. The evaluation method starts from the specific clinical use purpose of the implant interventional medical instrument to be evaluated and the anatomical structure constraint of the implantation position; user requirements are defined in a user domain, various properties including biomechanical properties, material properties and biological properties, structural design parameters and material design parameters are decomposed and refined in a functional domain,and ethical constraints are formed. The change conditions of various performance indexes of the implantable interventional medical instrument under the influence of different design parameters are obtained through testing methods such as mechanical testing, material science testing and biocompatibility testing; on the basis, a design matrix capable of quantitatively reflecting the change relationbetween the design parameters and the performance of the implantable interventional medical instrument is obtained, the design of the implantable interventional medical instrument is evaluated from the perspectives of the information amount and the coupling degree, and the evaluation result can provide a basis for optimization of the design of the implantable interventional medical instrument.

A design method for planting the rupicolous plant on the artificial bank earth in garden features that the cobbles with irregular shapes and different sizes are spread on the specially designed matrix to form an artificial bank earth and the rupicolous plants are planted on it.

The invention relates to the technical field of tunnel positioning, and discloses an underground tunnel-oriented ridge regression two-dimensional positioning method and system. The method comprises the following steps: firstly, receiving signal arrival time from different positioning base stations through a positioning label to obtain a TDOA value; further determining an observation vector and a design matrix of the least square problem according to the position of the base station; calculating a condition number after the transpose of the design matrix is multiplied by the design matrix to estimate the noise sensitivity, and setting a proper ridge regression coefficient according to the condition number; and finally, determining a position vector, and determining coordinates of the positioning label according to the position vector. Through the above steps, the method overcomes the common noise sensitivity of TDOA positioning in an underground tunnel environment, and greatly improves the positioning precision and stability.

The invention discloses a calculating method for the geometric accuracy factor minimum value of a multi-mode receiver system layer. The calculating method comprises the steps that a design matrix of the multi-mode receiver system layer is built; a correlation matrix is processed through a block matrix inversion method; a calculating model of the geometric accuracy factor of the system layer is built; diagonal elements of the relevant portion are calculated correspondingly; and a calculating model of the geometric accuracy factor minimum value of the multi-mode receiver system layer is built. On the basis of the block matrix inversion principle, the correlation matrix is subjected to blocking according to the practical significance of a designed matrix, the diagonal elements of the relevantportion are calculated through the block matrix inversion method and an adjoint matrix, finally the geometric accuracy factor minimum value of the multi-mode receiver system layer is obtained, and the calculating process of the geometric accuracy factor minimum value of the system layer is simplified effectively.

The invention provides a sparse superposition code design scheme based on hybrid iteration power distribution, and the scheme specifically comprises the steps: carrying out the sparse representation of an input real value signal through employing a one-hot encoding mode, and enabling each group to have only one non-zero value; then, assigning the non-zero value in each group of the sparse messagevectors by using the hybrid iteration power distribution scheme provided by the invention; then performing linear combination of corresponding groups with the random Hadamard design matrix to form corresponding code words; secondly, inputting the formed code words into an AWGN channel, and adding Gaussian white noise with the mean value of 0 and the variance of 2 into the code words; and carryingout iterative estimation operation on the output noisy code word through an AMP decoding algorithm to obtain an estimation value of the message vector; setting the maximum value corresponding bit in each group of the obtained estimation values as a preset value obtained by power distribution, and setting the rest bits as 0 to obtain a reconstruction value of the original signal. An experiment is carried out by using spyder3, and a needed python third party package comprises numpy, pyfht and the like.