42 results about "Square matrix" patented technology

The invention relates to the technical field of heat conversion flow distribution, in particular to a self-adaptive flow distribution method and device for a double-pipe heat exchanger, and the method comprises the following steps: S1, equipment assembly: installing a plurality of groups of flow pipes in the middle of a shell, and installing two groups of flow adjusting assemblies on one side of the input end of each of the plurality of groups of flow pipes; s2, arranging a plurality of groups of flow tubes in the shell in an annular matrix or a square matrix, establishing a plane-coordinate system by taking the center of the matrix as a coordinate origin, and numbering and marking each flow tube; s3, flow grade judgment; according to the scheme, the flow grade of the flow pipe is determined, the higher the flow grade is, the faster the flow speed is and the larger the flow is, then the flow channel area of the high-grade flow pipe is adjusted to decelerate and reduce the flow, the redundant quantity is distributed to the low-grade flow pipe, and it is guaranteed that the flow speed and the flow difference value of all the flow pipes are smaller than the set threshold value through comprehensive adjustment. Therefore, the effect of reasonable flow distribution is achieved, and the thermal conversion efficiency is improved.

The invention provides a distribution line fault positioning precision self-correction method based on a digital twin model, and the method comprises the steps: independently constructing a virtual branch of a parallel admittance structure for each phase for each measurement point node of the digital twin model, and initializing the admittance vectors of all virtual branches; obtaining a three-phase current measured value of the distribution line, a three-phase voltage amplitude measured value and a power measured value of each measuring point node, and a corresponding power theoretical value in the digital twinborn model; obtaining a residual vector between the power measured value and the power theoretical value; assigning a division result of the residual vector and the voltage amplitude square matrix to an admittance vector; and acquiring the virtual power of each virtual branch according to the admittance vector, and correcting the admittance parameter of the distribution line in the digital twinborn model according to the current measured value. According to the method, the digital twin model can continuously and automatically track parameter changes of a physical system through a closed-loop correction mechanism, and the precision and reliability of fault positioning based on the model are ensured.

The invention provides an energy storage system black start method and system based on hierarchical control, and the method comprises the steps: carrying out the secondary frequency modulation and voltage regulation of a collected common voltage at a station control layer, so as to obtain a regulation variable, and transmitting a starting instruction, an off-grid mode enabling instruction and the regulation variable to a square matrix layer; a master matrix and a slave matrix are determined, the master matrix is controlled to perform black start, the master matrix is converted into an off-grid state after black start succeeds, and the black start of the master matrix comprises power loop control and voltage outer loop control performed by the master matrix and current inner loop control of an energy storage converter layer; controlling in-station alternating current bus voltage from grid connection of the square matrix to construction of the main square matrix, and converting into an off-grid state after grid connection is completed; and after receiving the off-network state information of the master square matrix and the off-network state information of the slave square matrix, the station control layer is switched into an off-network state, and the black start of the energy storage system is completed. According to the method, hierarchical control is carried out on the system, the advantages of serial recovery and parallel recovery are combined, and the stability of the system is ensured while the black start speed of the system is increased.

Matrix multiplication operations can be implemented, at least in part, on one or more tensor cores of a parallel processing unit. An efficiency of the matrix multiplication operations can be improved in cases where one of the input operands or the output operand of the matrix multiplication operation is a square matrix having a triangular data pattern. In such cases, the number of computations performed by the tensor cores of the parallel processing unit can be reduced by dropping computations and / or masking out elements of the square matrix input operand on one side of the main diagonal of the square matrix. In other cases where the output operand exhibits the triangular data pattern, computations can be dropped or masked out for the invalid side of the main diagonal of the square matrix. In an embodiment, a library implementing the matrix multiplication operations is provided.

An LDPC-code-based communication method and communication apparatus. In the method, a device can perform encoding or decoding on the basis of an LDPC matrix, wherein the LDPC matrix is determined on the basis of an LDPC base matrix, the base matrix comprises an extended check matrix, the extended check matrix is a lower triangular matrix with diagonal elements all being 1, the extended check matrix comprises a plurality of lower triangular matrices, diagonal elements of each of the plurality of lower triangular matrices are located on the diagonal of the extended check matrix, the positions of the diagonal elements of the plurality of lower triangular matrices do not overlap, and the elements in the extended check matrix other than the diagonal elements of the extended check matrix and the diagonal elements of the plurality of lower triangular square matrices are all zero elements. The method can increase the column weight of a check region, reduce the occurrence probability of an error plane, and also support retransmissions at some specific columns, and thus the method is compatible with an IR-HARQ scheme of 5G LDPC codes.

The invention relates to the technical field of planar motors, in particular to a stator unit for a moving magnetic planar magnetic suspension motor and the motor. The stator unit comprises a plurality of basic units, the first coil unit, the second coil unit and the position sensor unit in each basic unit form an n * n square matrix, and n is an integer greater than 1; the first coil units and the position sensor units on any odd-numbered rows or odd-numbered columns of the n * n square matrix are alternately arranged; the position sensor units and the second coil units are alternately arranged on any even-numbered rows or even-numbered columns of the n * n square matrix, and the coil direction of the first coil units is perpendicular to the coil direction of the second coil units. The motor comprises the stator unit, and the stator unit and the motor break through the constraint of a traditional cable, replace a laser interferometer scheme of a traditional moving-magnet planar motor, and simultaneously support the pose measurement of a plurality of rotors in a single oversized working plane.

The invention discloses a material array arrangement method, which comprises the following steps of: recording the time when materials pass through a fixed point through a capacitive sensor on an incoming material conveying belt, and calculating the real-time position of subsequent movement of the materials in combination with the speed of the conveying belt; a clamping device on the Z axis of the XYZ three-axis mechanical arm is used for clamping the materials and lifting the materials away from the incoming material conveying belt when the real-time positions of the materials are calculated to be within the clamping range; the materials are clamped and lifted according to the number of the materials in the set column direction, after a column array is completed, the column materials are placed on a conveying belt at equal intervals through a three-axis mechanical arm to form a square matrix, the conveying belt moves the square matrix to the next array position, and a new square matrix is generated at the array discharging position; a material matrix arranged at the array feeding position is conveyed to the material taking position through a conveying belt, and linkage of conveying and material taking is accurately controlled by a PLC through signals of a photoelectric sensor. According to the invention, through parameterized management of clamping stations and cyclic counting of array rows and columns, ordered arrangement of material matrixes of different specifications is realized, and the device is suitable for various working scenes.

The invention discloses a die clamping mechanism of a die-casting machine, which comprises a fixed die plate, a movable die plate and a tail plate, the fixed die plate, the movable die plate and the tail plate are sequentially connected through a pull rod assembly, the fixed die plate is fixedly connected with the pull rod assembly, and the movable die plate and the tail plate are movably connected through a bent elbow assembly. The pull rod assembly comprises a first rod set around the first flow guide groove and a second rod set around the second flow guide groove, all rod pieces of the first rod set are arranged in a square matrix mode with the center of the first flow guide groove as the center, and all rod pieces of the second rod set are arranged in a square matrix mode with the center of the second flow guide groove as the center. The invention relates to the field of die-casting dies, and aims to reduce the size and weight of equipment by optimizing the layout of pull rods and introducing a double-injection function.

The invention relates to the technical field of oil and gas exploration and development, and particularly discloses a robust mixed norm constraint high-dimensional seismic data reconstruction method and device, and the method comprises the steps: carrying out the Fourier transformation of five-dimensional data along a time axis, and obtaining the Fourier transformation data; unfolding 4D data of each frequency slice of the Fourier transform data along a mode-(m, n) to obtain a near-square matrix; a non-convex Frobenius / Nuclear mixed norm and L1 norm combined regularization constraint is applied to the near-square matrix, and a target functional of observation seismic data and reconstruction data is constructed; and alternately solving the target functional by adopting an ADMM algorithm, and carrying out inverse Fourier transform on a solving result along a frequency axis. According to the method, a high-dimensional tensor completion method under the robust Frobenius / Nuclear mixed norm constraint is introduced, accurate representation of nonlinear high-dimensional seismic data is achieved, the seismic data reconstruction and denoising precision is improved, and therefore regularized reconstruction and abnormal noise suppression of the seismic data with the low signal-to-noise ratio are achieved.

The present disclosure provides coding and decoding methods, a communication device, and a storage medium. The coding method, applied to a first transmission node, includes: determining a base matrix for low density parity check (LDPC) coding from a preset base matrix set; performing LDPC coding on original data according to the base matrix, so as to obtain coded data; transmitting the coded data to a second transmission node; wherein the base matrix includes two kinds of elements: an element indicating an all-zero square matrix, and an element indicating a cyclic shift of an identity matrix; the base matrix set includes at least two non-empty subsets of base matrices, and all base matrices in each subset of the at least two non-empty subsets of base matrices have a same coding rate.

1. The name of the design product: sound (square matrix). 2. The use of the design product: the design product is used for sound equipment. 3. The design points of the design product: the combination of shape and pattern. 4. The picture or photo that best indicates the design points: combination state diagram.

The invention discloses a high-order matrix operation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a plurality of target matrixes to be operated, dividing each target matrix into a plurality of square matrixes, and distributing each square matrix to a corresponding chip; a ping-pong transmission algorithm is adopted through a plurality of chips, and on-chip matrix multiplication is completed on the distributed square matrix; respectively corresponding multiplication results are transmitted to a target chip in parallel by adopting a ping-pong transmission algorithm through the plurality of chips; and performing accumulation operation on the multiple multiplication operation results through the target chip to obtain a target operation result corresponding to the target matrix. According to the technical scheme of the embodiment of the invention, the multiplication efficiency of the high-order matrix and the utilization rate of hardware resources can be improved.

The application discloses a kind of converter electromagnetic transient simulation parallel optimization method, equipment and storage medium, belong to power system electromagnetic transient simulation technical field, by merging significantly reduces unnecessary intermediate variable, shortens simulation calculation process;And since C, D matrix and Un, k matrix multiplication operation with Itotal of A, B matrix and multiplication are carried out simultaneously, and the calculation process between row and row is independent, increase the parallel degree of electromagnetic transient simulation calculation;Since parameter matrix does not change with the change of switch state, it is calculated and stored in advance in simulation initialization stage, only needs to consume register reading operation time in the calculation of single simulation step;Coefficient matrix A, B, C, D are all square matrix of scale Nb*Nb, the formula for calculating Ub and Ib is the same in structure, to lay the algorithm foundation for saving logic resources while improving the calculation efficiency of simulation program using pipeline and other hardware development techniques.

According to the square matrix movement control method provided by the invention, a movement instruction for a target square matrix is received, and the movement instruction comprises a target position, a target formation specification and a target orientation vector corresponding to the target square matrix; generating a target coordinate set according to the target position, the target formation specification and the target orientation vector; acquiring the current state of the target square matrix; sorting the target coordinate set based on the current state and a preset movement type, and determining an allocation sequence of each target coordinate; matching each unit in the target square matrix with a target coordinate in the target coordinate set to generate a corresponding relationship between the units and the target coordinate; and controlling each unit to move towards the corresponding target coordinate according to the corresponding relationship. Accurate control over square matrix movement can be achieved, and the efficiency and accuracy of square matrix transformation are improved through reasonable coordinate distribution and a matching algorithm.

Embodiments of the present disclosure provide a visual inertia fused localization method and a device. The method comprises: acquiring target observation data corresponding to a target state vector; constructing a first residual model according to the target observation data; projecting the residual model according to a first-order partial derivative matrix, to obtain a second residual model, the second residual model comprising a product term of a second-order partial derivative square matrix and a system state matrix, and a covariance term corresponding to a first observation noise term; and eliminating a three-dimensional coordinate error vector of the second residual model, to obtain a target residual model, and performing EKF updating on the target state vector according to the target residual model. The embodiments do not involve left null space calculation and projection, and QR decomposition dimension reduction calculation, thereby reducing the amount of calculation, and improving positioning efficiency.

The application provides a radionuclide activity calculation method based on a Padé approximation considering multistage decay, which comprises the following steps: S1, constructing a complete decay matrix containing all nuclides; step S1 comprises the following steps: S 11 , receiving a decay chain data structure as input, constructing an n*n square matrix A; S 12 , initializing all elements of the entire matrix to zero values; S 13 , setting the element of the ith row and the ith column to the negative value of the decay constant of the ith nuclide, that is, A(i,i)=; S 14 , traversing all possible nuclide pairs (i,j); S2, calculating the matrix exponential by using a matrix norm driven adaptive scaling algorithm combined with a Padé approximation matrix; S3, performing corresponding radionuclide activity calculation according to the calculation mode. The application avoids truncation error by constructing a complete decay matrix and considering the contribution of all nuclides. The adaptive scaling algorithm combined with the Padé approximation matrix is used to solve the rigid system matrix calculation, which effectively avoids numerical overflow and rounding error, and guarantees the stability of the calculation process.

The invention relates to a matrix aided training method and system, belongs to the technical field of matrix aided training, and solves the problem that a matrix aided training method and system for improving training efficiency is lacked in the prior art. Comprising the following steps: acquiring a real-time video stream corresponding to a square matrix; detecting the real-time video stream by adopting a trained square matrix detection model to identify the positions of all individuals in the video stream; performing density clustering based on the positions of all individuals to generate at least one candidate region; wherein one candidate area corresponds to one square team; performing virtual grid fitting on each candidate area to form an optimal virtual grid; predicting the position of the matrix individual at the next moment based on the position of each grid point in the optimal virtual grid fitting result, the moving direction of the grid point and the number of pixels; and prompting the correct position of each individual based on the predicted position of each individual. The matrix aided training method and system are realized.

The predictive uncertainty of a generative machine learning model can be estimated. The generative machine learning model can be a large language model or a large multimodal model. A data element can be input into the generative machine learning model. The generative machine learning model can generate outputs from the data element. Latent embeddings for the outputs can be extracted from the generative machine learning model. A covariance matrix with respect to the latent embeddings can be computed. The covariance matrix can be a two-dimensional matrix, such as a square matrix. The predictive uncertainty of the generative machine learning model can be estimated using the covariance matrix. For example, the matrix entropy of the covariance matrix can be determined.Matrix entropy can be an approximate dimension of a latent semantic manifold spanned by the outputs of the generative machine learning model and can indicate the predictive uncertainty of the generative machine learning model.

The application discloses a kind of municipal building underground drainage pipe network operation risk assessment method and system, it is related to data processing technical field, the application constructs real-time hydraulic situation three-dimensional tensor by collecting the liquid level and flow rate data of monitoring point, input hydraulic topology expansion gate network obtains future hydraulic situation prediction tensor;Target analysis pipe section is spliced in time and space grid, and the internal physical field characteristics are solved in combination with prediction tensor and physical constraint, and the pipe wall mechanical response characteristic matrix is obtained;Based on the matrix, the pipe wall structure fatigue time-varying state transition chain is constructed, and the transient structure damage risk probability is obtained by Markov chain iterative calculation;Combined with real-time physical field characteristics, construct cross-modal cross attention operator to obtain weight square matrix;Finally, the comprehensive risk score is calculated in combination with the two. The application breaks through the blind area of traditional physical monitoring, realizes the continuous high-precision analysis of microscopic stress state and the dynamic quantification of transient fatigue fracture risk.

The application provides a bearing fault diagnosis method based on correlation entropy and short-time Fourier transform, and the diagnosis method is as follows: step 1, collecting a vibration signal x(i), the sampling length is N, the signal x(i) is an N*1 column vector, and the kernel matrix M of the signal is calculated x ,M x (i,j) = K[x(i),x(j)], K(·) is a kernel function, e (·) is a natural exponential function, sigma is a kernel length, i, j = 1, 2, 3,..., N, M x is an N*N square matrix, since the traditional short-time Fourier transform is susceptible to interference noise, and the bearing outer ring fault characteristic frequency and the system inherent vibration frequency are coupled with each other, it is difficult to effectively extract the bearing outer ring fault characteristic information under the noise interference, compared with the traditional short-time Fourier transform method, the application can effectively suppress the Gaussian noise and non-Gaussian noise in the signal, has the self-adaptive noise reduction performance, and can highlight the bearing fault characteristics.

The application provides a hyperspectral image shadow enhancement method, equipment and a storage medium, wherein the method comprises the following steps: acquiring a hyperspectral image dataset; normalizing the hyperspectral dataset and extracting a shadow part by using a mask; continuously sampling and unfolding the shadow part image in a row scanning manner in a spatial dimension into a one-dimensional signal, sequentially passing through a bistable nonlinear system for dynamic stochastic resonance; re-composing a square matrix from the sequence output by the system in a row scanning manner and then normalizing; continuously sampling and unfolding the obtained shadow part data in a column scanning manner into a one-dimensional sequence, and inputting the one-dimensional sequence into the bistable nonlinear system again for dynamic stochastic resonance; re-composing a square matrix from the output sequence in a column scanning manner and then normalizing, and fusing the normalized square matrix with a non-shadow area of the original hyperspectral image. The application utilizes the scanning and sampling modes in two different directions of rows and columns, fully utilizes the correlation of the pixels in the spatial dimension, retains more image information, and achieves better image enhancement effect.

The invention discloses a municipal building underground drainage pipe network operation risk assessment method and system, and relates to the technical field of data processing.The method comprises the steps that a real-time hydraulic situation three-dimensional tensor is constructed by collecting liquid level and flow velocity data of monitoring points, and a hydraulic topology expansion gating network is input to obtain a future hydraulic situation prediction tensor; carrying out space-time grid segmentation on the target analysis pipe section, solving internal physical field characteristics in combination with a prediction tensor and physical constraints, and obtaining a pipe wall mechanical response characteristic matrix; constructing a pipe wall structure fatigue time-varying state transfer chain based on the matrix, and obtaining a transient structure damage risk probability through Markov chain iterative calculation; constructing a cross-modal cross attention operator in combination with real-time physical field features to obtain a weight matrix; and finally calculating a comprehensive risk score by combining the two. According to the method, a traditional physical monitoring blind area is broken through, and continuous high-precision analysis of the microscopic stress state and dynamic quantification of the transient fatigue fracture risk are achieved.

A method including the following steps: establishing a matrix with i rows and j columns which includes i×j elements, where each element in the matrix is a square matrix with n rows and n columns, wherein i≥2, j≥2, n≥3, and i, j, n are all positive integers; using a magic square to label the numbers 1, 2, . . . , n2, on n2 coordinates of each square matrix respectively, so that a sum of the numbers labeled on n coordinates of each row, column, and both diagonals of each square matrix is equal ton⁡(n2+1)2;dividing a plurality of LED dies into n2 wavelength segments by wavelength, with each wavelength segment having i×j LED dies; and placing the i×j LED dies of the same wavelength segment on the coordinates labeled with the same number in the square matrix.

The application provides a Hamiltonian simulation method, device and storage medium, and solves the technical problem that it is difficult to simulate a Hamiltonian in the related art, so as to realize evolution simulation of the Hamiltonian. The Hamiltonian simulation method comprises the following steps: determining a Hamiltonian H to be simulated; wherein H is in the form of a square matrix and is independent of time; calculating e iA ; wherein A is a square matrix and A = -Ht, t is a constant, and the calculation result is in the form of a square matrix; decomposing the calculation result into a set of a limited number of quantum gates; and constructing a quantum circuit according to the set of the limited number of quantum gates to perform simulation.

The invention relates to the technical field of communication encryption, and provides an efficient search modeling method and device for separable propagation of a linear component, a medium and a product, and the method comprises the steps: converting a linear transformation matrix into an origin matrix, and calculating a matrix based on the origin matrix; limiting the relationship between the input separability and the output separability, and determining a logic expression in which the matrix is a square matrix, any column vector is a non-zero vector and any row vector is a non-zero vector; and converting the logic expression into a constraint condition through grammar, and calling a solver for solving. The method is suitable for separable propagation modeling of a complex linear layer, extra variables do not need to be introduced, compared with a disclosed modeling method, the efficiency of the method can be improved by 4-5 times, and the cryptographic algorithm analysis efficiency is remarkably improved.

A predictive uncertainty of a generative machine learning model can be estimated. The generative machine learning model can be a large language model or a large multi-modal model. Data can be input into the generative machine learning model. The generative machine learning model can generate an output from the data. A latent embedding of the output can be extracted from the generative machine learning model. A covariance matrix can be computed over the latent embedding. The covariance matrix can be a two-dimensional matrix, such as a square matrix. The predictive uncertainty of the generative machine learning model can be estimated using the covariance matrix. For example, a matrix entropy of the covariance matrix can be determined. The matrix entropy can be an approximate dimensionality of a latent semantic manifold spanned by the output of the generative machine learning model and can be indicative of the predictive uncertainty of the generative machine learning model.

Example parameter optimization methods and apparatus are described. In one example parameter optimization method, a data processing device obtains a Kronecker factor matrix that is used to indicate a higher-order information matrix of a neural network model, and segments the Kronecker factor matrix to obtain a plurality of square matrices. The obtained plurality of square matrices are submatrices of the Kronecker factor matrix, and main diagonal lines of the plurality of square matrices each are in a one-to-one correspondence with a part of a main diagonal line of the Kronecker factor matrix. Then, the data processing device adjusts a parameter of the neural network model based on the obtained plurality of square matrices.

The application belongs to the technical field of signal processing. The application provides a complex scene DOA estimation method based on matrix reconstruction. The method comprises the following steps: obtaining a received signal, and obtaining a signal covariance matrix of the received signal according to the received signal; performing eigenvalue decomposition on the signal covariance matrix to obtain a signal characteristic vector corresponding to a signal subspace; extracting a maximum signal characteristic vector corresponding to a maximum eigenvalue of the signal characteristic vector, and performing element rearrangement on the maximum signal characteristic vector to obtain a reconstruction matrix; and performing DOA estimation based on the reconstruction matrix and in combination with a Root-MUSIC algorithm based on polynomial reconstruction to obtain a DOA estimation value. The correction matrix in the embodiment of the disclosure is a square matrix, and the DOA estimation under a complex scene can be realized only by performing eigenvalue decomposition. Moreover, the success probability is high, the signal-to-noise ratio threshold is low, and the estimation performance is good.