143 results about "Vector distribution" patented technology

The invention, which belongs to the power converter control field of the driving motor, relates to a method for controlling neutral point voltage balance of a three level converter, wherein the method is based on a voltage space vector pulse width modulation technology and is used for controlling a neutral point clamped three level inverter. The method comprises the following steps that: step one, an action time distribution factor of a redundant small vector is obtained; step two, an area where control on a neutral point potential balance can be realized by a nearest three vector (NTV) method is obtained; step three, an NTV action area in a mixed voltage space vector pulse width modulation method is obtained; step four, a switching condition of the mixed voltage space vector pulse width modulation method is designed; and step five, the mixed voltage space vector pulse width modulation method with a proportion of pm is realized. According to the invention, a redundant small vector distribution factor is employed to design a switching condition of a mixed voltage space vector pulse width modulation strategy; therefore, an essence of a space vector modulation method is embodied; themethod can be used flexibly and used variables are less, so that it is beneficial for analysis and development application.

The invention discloses a method for modulating space vectors of double inverters, and belongs to the field of inverter control. The method is used for solving the problem that in the prior art, when basic space voltage vectors which do not generate common-mode voltage are selected to modulate voltage, the utilization of direct-current bus voltage is low. The method comprises the steps that a space voltage vector plane is divided into six sectors, a reference voltage vector is divided into two vectors, one vector is equivalent to the operation that one of the double inverters is clamped in a specific switching state, and the other vector is modulated by the other inverter in six sub-sectors of the inverter to be achieved; modulation is conducted within the maximum voltage range, an equivalent zero vector distribution factor is introduced when the vectors are synthesized, the common-mode voltage generated in the process of synthesis of the basic voltage vectors is counteracted with common-mode voltage generated by a zero vector, and common-mode voltage of the double inverters is restrained.

The invention discloses a distributed driving electric automobile torque vector control method based on double-layer control. The control method comprises the following steps: establishing a vehicle dynamic model, and calculating a tire slip rate by utilizing a Dugoff tire model; establishing a torque distribution controller based on the vehicle dynamic model and the Dugoff tire model, wherein thetorque distribution controller comprises an upper controller and a lower controller; calculating the current drive torque of each driving wheel of the vehicle by the upper controller according to body yawing angular velocity; calculating a compensating torque needed for remaining an ideal slip rate of each driving wheel by the lower controller according to the ideal slip rate of each driving wheel serving as a control objective, and further performing compensation distribution on the driving torque, so that an actual torque is output to the driving wheel, and torque vector control is completed. According to the control method disclosed by the invention, the automobile torque can be effectively subjected to vector distribution, the vehicle driving stability and smoothness are improved, theoperating burden of the driver is obviously decreased, and the driving safety is improved.

The invention provides a gray scale and geometric information combined medical image registration method and relates to the technical field of medical image processing. Grey scale values are the values on all pixels/voxels in digital medical images and distributed as image gray scale information; and normal vectors represent unit vectors of geometric normal of all points, represent the gray scale information of the medical images by adopting the grey scale value distribution, represent the geometric information of the medical images by adopting the image normal vector distribution, and finally measure the similarity between the gray scale and the geometric information quantity by adopting mutual information. According to the invention, the conventional mutual information method and gradient solution method are adopted, and the obtained mutual information values are added together, so that easy implementation can be realized in engineering application; the advantages of both the gray scale and the geometric information are fully utilized to ensure that less local extreme points fall in the registering process, and better robustness is achieved; and tests show that the method provided by the invention can acquire higher precision.

An off-axis aspheric lens detection method based on fringe reflection includes: firstly, building a world coordinate system around the center of an off-axis aspheric lens and acquiring the relation among the world coordinate system, a display coordinate system and a camera coordinate system; secondly, utilizing the camera to take pictures of projections, reflected from a measured lens, of sine periodic fringe patterns displayed on a display screen during measuring; and thirdly, obtaining phase distribution according to the phase shift technology and the phase unwrapping technology, obtaining normal vector distribution of the measured off-axis aspheric lens according to the relation among the coordinates; and finally, rebuilding surface form height information of the measured off-axis aspheric lens in an integral manner. In addition, a measuring system utilizing the off-axis aspheric lens detection method based on fringe reflection consists of the measured lens, the display screen and a CCD (charged coupled device) camera unit.

The invention discloses a method for optically measuring shear stress of boundary layers of flow fields, and relates to the technical field of wind tunnel tests. The method includes main steps of (1), anodizing and blackening wind tunnel test models; (2), smearing liquid crystal coatings; (3), acquiring relation curves of colors of the liquid crystal coatings and shear force; (4), mounting 6 test cameras; (5), simultaneously photographing the liquid crystal coatings by the 6 test cameras to obtain images of the liquid crystal coatings; (6), removing position difference of the images photographed by the different test cameras; (7), carrying out distortion removal processing; (8), correcting identical colors of the images photographed by the various cameras; (9), acquiring model surface shear stress vector distribution results. The method has the advantages that the obtained images are clear and visual, and the method is high in measurement accuracy and sensitivity and can be used for displaying the flow fields of wind tunnels, diagnosing transition and quantitatively measuring model surface friction drag.

The invention discloses an intelligent vehicle speed prompting driving system based on a mobile phone. The intelligent vehicle speed prompting driving system comprises the mobile phone on a vehicle and a server. According to the work processes of the mobile phone and the server, the GPS and accelerated speed information of the mobile phone on the vehicle are collected in a crowdsourcing mode, and data are transmitted to the server through a wireless network at certain time; the positions of road junctions with traffic lights are determined through an accelerated speed vector distribution map and a GPS path; a road is represented by using a secondary B spline; the accelerated speed events of each road junction are collected in a crowdsourcing mode, and the time tables of the traffic lights of the road junctions are deduced; the optimal speed is prompted for the vehicle according to a generated map and the time tables of the traffic lights of the front road junctions, and the situation that the vehicle meets a red light at the road junctions and stops is avoided as far as possible. On one hand, the map is generated though a new method, and on the other hand, the deduction of the time tables of the traffic lights at the complex road junctions is solved through a road junction structure.

The invention discloses a texture force touch sensing method based on single image fractional order processing. The method includes the steps that a fractional differential algorithm is utilized, the order gamma of the fractional differential algorithm is adjusted dynamically, an enhanced texture image is obtained, original image texture information is extracted from the enhanced texture image, and multi-scale analysis of the detail texture and the edge contour of different rates of gray level changes is achieved from the point of image airspace gray level change features; force touch rendering is carried out on the virtual surface texture, and tangential force frictional damping coefficients of a force touch sensing model are controlled on the basis of gradient vector distribution of the enhanced texture image; obtained resultant force is calculated and fed back to an operator through a hand controller. By the utilization of sensibility of fractional differential to detail information, the sense of reality of texture force touch sensing is enhanced by manually selecting orders, extracting texture information in a two-dimensional image signal from different scales and converting the texture information into frictional damping control.

An adaptive search range method for motion/disparity estimation is provided in multi-view video coding (MVC) technology. The method uses a initial search range as a first search range, perform an estimation flow for first blocks in a first frame to obtain vector distribution, and obtain at least a first candidate search range in accordance with the vector distribution. The first candidate search range is selected as a second search range to perform estimation flow for second blocks in a second frame, and an estimation vector according to the estimation flow is obtained and provided as video coding.

The invention discloses a method for automatically cutting a granular object in a digital image, belonging to the technical field of digital image processing. The method comprises the following steps of: firstly, separating an object from a background by applying an automatic threshold method by aiming at characteristics such as gray level, structural distribution, geometry and the like of the granular object in the digital image, particularly a microscopic image; then, calculating a gradient vector field of the object, and searching a key point in the gradient vector field, wherein the idealkey point has corresponding gradient vector distribution in eight neighborhoods, the gradient value of the key point is zero, and the acquired key point is used as the center of each granular object;next, defining a new effective energy function based on the gray level and a space position so as to calculate a direction gradient to replace the traditional gray level gradient; and finally, searching the boundary of the granular object by applying an active contour model. By using the method, the aggregated granular object can be accurately and effectively cut, particularly, a great number of adhered or overlapped micro-grains exist in a biomedical microscopic image, and therefore, help is provided for the image analysis and identification.

The invention discloses a flow field characteristic experiment device of a combustion chamber of a hydrogen-burning gas turbine. The flow field characteristic experiment device comprises three pairs of bluff body arrangement combustion chamber experiment pieces, a particle image velocimetry (PIV) test platform frame, a PIV test system and a pressure difference and environment temperature test system. The three pairs of bluff body arrangement combustion chamber experiment pieces comprise a front lateral plate, a front bluff body, a back bluff body, an affiliation mechanism of the back bluff body, a front top plate and a back top plate. A tracer particle injection pipe jack is arranged on the front lateral plate, an inlet speed probe jack is arranged on the front top plate, and an outlet speed probe jack is arranged on the back top plate. A tracer particle generator is connected with the tracer particle injection pipe jack on the front lateral plate. The device adopts the PIV test technology, selects reasonable combustion chamber position, reasonably mingles tracer particles into a main gas flow under the effects of the tracer particle generator and a tracer particle injection pipe, enables tracking particles in the test area to be evenly distributed and conducts visualization experiment of cold state flowing speed vector distribution of three pairs of bluff arrangement combustion chambers.

The invention discloses an antagonistic cross-media retrieval method based on bilingual meaning space, which relates to the technical fields of pattern recognition, natural language processing, multimedia retrieval and the like. Including: feature generation process, bilingual semantic space construction process and antagonistic semantic space optimization process. The invention realizes that theoriginal image and the text information are kept to the maximum extent while the semantic gap is eliminated by establishing an isomorphic bilingual meaning space, namely a text subspace and an image subspace. And through antagonism training to optimize the distribution of isomorphic subspace data, mining the rich semantic information in multimedia data, and fitting the vector distribution of different modes in the semantic space under the condition that the categories remain unchanged and the modes can be distinguished. The method of the invention can effectively eliminate the heterogeneity ofdifferent modal information, realize effective cross-media retrieval, and has wide market demand and application prospect in the fields of picture and text retrieval, pattern recognition and the like.

The invention provides a cross-media information analysis and retrieval method, which comprises the following steps of: performing semantic integration processing on multimode information; performing expansion according to a probability latent semantic analysis model to obtain a multilayer continuous probability latent semantic analysis model for processing a continuous feature vector; learning the multilayer-continuous probability latent semantic analysis model by adopting an asymmetric learning method, and calculating the visual feature vector distribution of an image, the visual feature vector distribution of an audio and topic probability distribution; submitting a training set and a tested media object which serves as a retrieval example by a user, and calculating intra-mode and inter-mode initial similarity values of the image and the audio in the retrieval sample; constructing a propagation model, and updating the intra-mode and inter-mode similarity values according to the propagation model; and performing secondary retrieval according to the updated similarity values.

The invention relates to a method for building and digitally expressing grassland vegetation subtypes. The method comprises the following steps of (1) marking a searched area, and searching vegetation of acquired searching sample points; (2) calculating the plant species dominance degree of vegetation of each searching sample point; (3) according to the plant type and the plant species dominance degree of each searching sample point, judging whether the searching sample points have the same vegetation subtype or not, and when the vegetation subtype is changed, actually measuring the boundary coordinates of different vegetation subtypes to obtain the vector base map of each vegetation subtype in the searched area; (4) building an attribute data table of each vegetation subtype, and drawing the digital vector distribution diagram of each vegetation subtype by adjusting and overlapping transparency of diagram layers; (5) placing the obtained digital vector distribution diagrams of the vegetation subtypes together, and finally generating the digital vector distribution diagram which shows the vegetation subtype distribution in the searched area and the formation and distribution of dominant plants in the vegetation subtypes. The method is widely applied to practical applications, such as research on the evolution of grassland vegetation, agricultural production, grassland vegetation resource monitoring.

A method for calculating a local specific energy absorption rate (SAR) on basis of an electric parameter and the mass density of a segmented geometry of an object and a magnetic field vector distribution of a radio frequency (RF) antenna. The values of the electric parameter and the mass density are pre-determined values, while the magnetic field vector distribution is estimated by a magnetic field mapping method based on a magnetic resonance (MR) scan. The magnetic field mapping method based on a magnetic resonance scan can be a Bi mapping method. The invention also relates to a magnetic resonance system by means of which SAR calculation can be done in a relatively short period of time. The invention also relates to a computer program comprising instructions for calculating a local specific energy absorption rate (SAR) according to the above mentioned method. The SAR calculation used in the above mentioned method, system and program is done in the relatively short period of time and as such is practicable in a clinical setting.

The invention relates to a design analysis method for a gas turbine gas inlet device, which belongs to the technical field of gas turbines. The precise stimulating analysis is carried out on the working conditions of a local part and a section of a gas inlet device by adopting a computer stimulating technology. The method comprises the following steps of: (1) model establishment: firstly, primarily designing a two-dimensional viewgraph and carrying out ANSYS three-dimensional modeling on the part needing to be calculated and analysis under the condition of confirming the sizes of all parts of the two-dimensional viewgraph; carrying out finite element division of an analysis object according to the concrete technical requirement and precision requirement needing analysis and calculation of the analysis object so that a computer can analyze and calculate each unit by utilizing enough resources; and (2) loading analysis: A. strength and linear buckling analysis: containing stress distribution, displacement distribution and load factors in calculating results; B. flow field analysis: analyzing the static pressure, total pressure and speed flow line of a system and respectively obtaining a speed vector distribution picture, a static pressure cloud picture and a total pressure cloud picture of a typical section.

The invention relates to big data, and provides an abnormal case identification method based on artificial intelligence which is applied to a platform server. The method comprises the steps: acquiringdiagnosis and treatment project data corresponding to the same disease, taking diagnosis and treatment time as a unit, and taking the diagnosis and treatment project data as words in a text to conduct text construction, wherein the constructed text comprises the diagnosis and treatment project data of different users of the same disease in the corresponding diagnosis and treatment time period; performing topic model construction on the text to obtain corresponding topic vector distribution, wherein each topic vector distribution is corresponding to one topic and comprises a plurality of wordidentifiers and corresponding word weights, and the topics have different user topic weights for different users; aligning the topic vector distribution corresponding to each text and performing clustering to obtain diagnosis and treatment item categories corresponding to each user at different time to form a diagnosis and treatment sequence corresponding to the user; and inputting the diagnosis and treatment sequence corresponding to each user at different diagnosis and treatment time into the process mining model to obtain a standard clinical path corresponding to the disease type so as to identify the abnormal case.

The invention discloses an adaptive visual positioning method of heart blood flow vortex movement based on color doppler image information. The method comprises the following steps of according to the distribution conditions of blue areas and red areas of a color doppler ultrasound heart blood flow image, determining the approximate area of the blood flow vortex movement; in the rough positioning area, determining the accurate boundary of the blood flow vortex movement area; in the accurate positioning area, obtaining a speed vector distribution chart of blood flow vortex movement; according to the speed vector distribution chart of blood flow vortex movement, obtaining a plane streamline chart of blood flow vortex movement; according to the plane streamline chart of blood flow vortex movement, judging the volume, shape and position information of blood flow vortex movement. The method has the advantages that the defect of objective selection of interested areas of the existing method is overcome, the precision of visual positioning of blood flow vortex movement of a heart flow filed is greatly improved, the solid foundation is laid for the accurate quantitative evaluation of blood flow vortex movement of the heart flow filed, and the accurate mechanical data support is also provided for the accurate evaluation of the heart mechanical function.

The invention discloses an ultraviolet light non-line of sight (NLOS) communication directional access method based on angular sensing. The method comprises the steps of: firstly, setting the value of a receiving elevation angle to be 90 degrees, and calculating the coverage area of ultraviolet light NLOS communication according to a divergence angle and a launching elevation angle of an emission node; then designing a node structure of the ultraviolet light NLOS communication, and establishing an interface elevation angle neighbor list; then updating a directional network vector distribution table; then judging whether a channel is idle or not; performing channel reservation; and finally, sending and receiving data to complete the direction access of the ultraviolet light NLOS communication. According to the method disclosed by the invention, an optimal launching elevation angle and an optimal interface number can be selected by an ultraviolet light communication node during channel access according to the interface elevation angle neighbor list, for data transmission. Compared with an access algorithm of a static appointed transmission elevation angle, the method disclosed by the invention has the advantages of reducing the probability in interference node generation, increasing communication transmission distance, reducing a hop count needed by communication, and providing relevant guidance for the node structural design with adoption of an ultraviolet light NLOS communication manner.

The invention discloses a scientific and technological paper clustering analysis method based on variational diagram auto-encoder and K-Means, which comprises the following steps of: constructing a citation network G=(V, E, F) by utilizing existing scientific and technological paper data, and constructing a variational diagram auto-encoder consisting of an encoder and a decoder according to an adjacent matrix A of a citation relationship between papers and a characteristic matrix F of paper keyword attributes, taking minimization of distance measurement between a reconstructed adjacency matrixand an original adjacency matrix A and divergence of node representation vector distribution and normal distribution as targets, training in an unsupervised mode to obtain multi-dimensional Gaussiandistribution, and sampling from the distribution to obtain a low-dimensional embedded vector z of a node; and then clustering the low-dimensional embedded vector z by using the K-Means algorithm to obtain a division result of the science and technology paper, and performing two-dimensional visual display after dimension reduction by using a tSNE algorithm. According to the method, the accuracy ofscientific and technological paper clustering analysis is improved, and the calculation cost of analysis is reduced.

The invention belongs to the technical field of radar information processing, particularly relates to a radar information processing method used for restraining false-alarm objects of rotary blades of wind driven generators on the ground during areal detection of the radar. The radar information processing method includes the steps of 1, acquiring a pulse transmitting frequency of the radar, accumulating pulse counts, and calculating a Doppler frequency through a conjugated product vector; 2, determining a main clutter position on a Doppler spectrum according to the Doppler frequency of the main clutter; 3, acquiring angular spin rates and the largest strength of the rotary blades of the wind driven generators, and calculating width of spectrum filter corresponding to the tail ends of the rotary blades of the wind driven generators; 4, acquiring distance gate numbers and power of detecting points, average noise power of the frequency spectrum and semi-diameter of broadening of the main clutter, and according to the judged threshold coefficient, acquiring a two-value vector distribution of the signal local power, with the main clutter as the center and the spectrum broadening of the rotary blades as the semi-diameter, on a single distance gate; 5, judging and removing the false alarms.