16results about How to "Avoid heavy computation" patented technology

The invention provides a health detection method and device for industrial equipment and a storage medium, and is applied to the technical field of industrial equipment, and the method comprises the steps: obtaining first feature data corresponding to to-be-detected industrial equipment; wherein the first feature data is determined based on vibration data corresponding to the to-be-detected industrial equipment; inputting the first feature data into a health detection model to obtain a health detection result output by the health detection model; wherein the health detection model is obtained by performing unsupervised learning and semi-supervised learning training on an initial model, and the initial model is a type of support vector machine. The AI-based data-driven development mode has relatively high development efficiency, so that the problem of relatively low development efficiency of a traditional expert system is solved. Besides, in the process of obtaining the health detection model through the catenary, an unsupervised learning mode and a semi-supervised learning mode are combined, so that the confidence coefficient of the label is improved, and the cost of performing health examination on the industrial equipment can be reduced.

The invention discloses a crankshaft point cloud semantic segmentation method. The method comprises the steps of performing voxelization downsampling on a crankshaft point cloud model and establishing a Kd-Tree topological structure; calculating normal lines under two different support radiuses, adjusting the direction of the normal lines, and calculating a normal line differential (DoN) characteristic according to a normal line difference; obtaining the number of point neighborhoods with a small support radius, and calculating point cloud curvature and FPFH features; defining an initial target area by taking a point with the minimum number of neighborhood points as an initial seed point; searching a k neighbor point set of the seed points, taking an FPFH feature threshold value and a curvature threshold value as region growth criteria, and performing semantic judgment on a segmented region by using a DoN feature threshold value; when the neighborhood point meeting the growth criterion cannot be found, completing the growth of the current region and continuing the growth of the next region until the whole point cloud is traversed; and completing crankshaft point cloud semantic segmentation according to the number of the marked region points. The method is low in complexity, can accurately complete crankshaft point cloud semantic segmentation, and improves the universality of a crankshaft point cloud segmentation algorithm.

This invention provides a radiometric calibration and temperature inversion method for a cooled infrared thermal imager, comprising: constructing a temperature calibration device, acquiring multiple calibration temperature points and their corresponding blackbody radiance to obtain corresponding datasets, acquiring two sets of response image grayscale value datasets to obtain a radiometric calibration model; plotting response curves to simultaneously solve for unknown parameters in the radiometric calibration model to obtain a near-range radiometric calibration model; using the modified radiometric calibration model for back-calculation to obtain the actual radiance of the target; pre-setting the constructed lookup table in the embedded processing unit of the infrared thermal imager to perform online temperature inversion output to obtain the absolute temperature of the measured target, and then outputting the absolute temperature of the measured target to a host computer display module. This invention can significantly reduce the calibration workload, enabling the infrared thermal imager to flexibly adjust the integration time and maintain high-precision temperature measurement in complex scenarios, significantly improving its engineering practicality and environmental adaptability.

The invention relates to the technical field of vehicle collision avoidance, and discloses a vehicle collision avoidance method, device and equipment, and the method comprises the steps: when a vehicle turns, preliminarily determining whether a target two-wheeled vehicle is on a driving track of the vehicle according to the predicted position information of the vehicle and the target two-wheeled vehicle at the first collision time and the second collision time, and if yes, determining that the target two-wheeled vehicle is on the driving track of the vehicle; if so, taking the two-wheeled vehicle as a preliminary risk target; determining an important risk target according to the first collision time and the second collision time of the initial risk target, the relative distance between the vehicle and the initial risk target, and the relationship between the initial risk target and the driving track of the vehicle when the initial risk target takes the collision avoidance action; and when the vehicle decelerates to avoid collision, determining a target deceleration according to a preset deceleration calculation rule, and determining whether to trigger automatic emergency braking according to the relationship between the target deceleration and the preset deceleration. According to the application, under the condition that one camera is arranged, whether the automatic emergency brake is triggered or not can be accurately judged.

The invention discloses a large-range collaborative navigation method for an extraterrestrial patroller, and relates to the field of computer vision, visual navigation and map splicing. The method comprises the following steps: performing significance analysis based on spectral residual for a right monocular camera, focusing on sight capture of a typical feature region, establishing a visual odometer model of inertial IMU pre-integration, and performing back-end optimization driven by a factor graph; establishing a plurality of extraterrestrial patroller similar region search models based on an improved bag-of-words model, introducing a similarity relative scoring function based on a norm form, setting a relative threshold value of a similarity suppression function, and obtaining an image overlapping region; according to the method, the calculation capability of an airborne computer is considered, a sparse feature point map and patroller pose splicing model is established, complete splicing of overlapped area maps is completed through splicing of small sparse feature point maps, and the splicing precision and the splicing efficiency are improved.

The invention relates to a VSP missing data reconstruction method and device based on a self-supervised deep learning model, and the method comprises the steps: inputting a VSP common detection point gather containing shot missing data into a deep convolutional neural network model; reconstructing the shot-lacking data by using a deep convolutional neural network model, and outputting a reconstructed complete VSP common detection point gather; wherein the deep convolutional neural network model is obtained by training in a self-supervised learning mode by adopting an original complete VSP common detection point gather in a target work area; according to the method, a self-supervised learning normal form is adopted, an existing complete gather data training model in a target work area is directly utilized, additional labeling or complex speed model construction is not needed, and the data preparation cost and dependence on prior knowledge are remarkably reduced; the trained model can perform end-to-end intelligent reconstruction on the input incomplete gather, the complete gather is quickly output, and the problems of complex iteration and large calculation amount in a traditional method based on a wave equation or matrix completion are avoided.

The invention belongs to the technical field of wind field feature analysis, and discloses a CFD-based complex mountain terrain wind field feature analysis method and system, and the method comprises the steps: obtaining mountain curved surface terrain data of a target site, and constructing a wind field model based on the mountain curved surface terrain data; setting a turbulence model, a boundary condition and a grid of the wind field model based on the wind field model; acquiring a plurality of feature observation points from the wind field model, and determining a change rule of the wind speed along with the height based on simulation data of the feature observation points; extracting wind speed and wind pressure characteristics of different characteristic observation points at different heights, determining a wind speed correction coefficient of a mountain curved surface based on a change rule of the wind speed along with the height, performing full-process simulation analysis on a complex mountain wind field through a wind field model, and converting mountain curved surface topographic data of a target field to obtain a mountain curved surface topographic data model; compared with traditional field measurement and wind tunnel tests, the cost is reduced, the method is not limited by weather and physical models, and continuous wind field information in the whole target area can be efficiently obtained.

The application discloses a kind of autonomous integrity monitoring methods for satellite navigation receiver, convolutional neural network is used in the present application and the slope weighted least square RAIM combined algorithm of optimization.The combined algorithm includes: first, the characteristic slope of satellite is calculated to select satellite, after eliminating the satellite with the largest characteristic slope, when the number of satellites is greater than or equal to 5, the optimized slope weighted least square RAIM algorithm is used, and the corresponding weight matrix is constructed based on the RAIM algorithm based on least square residual;When the number of satellites is less than 5, data regression learning prediction is carried out using convolutional neural network, to determine whether there is a faulty satellite.The present application has the function of more accurate monitoring whether satellite signal is faulty, to achieve fault monitoring and troubleshooting, so as to realize the autonomous integrity monitoring of satellite navigation receiver.

The invention belongs to the technical field of mechanical engineering magnetic liquid sealing, and particularly relates to a multi-scale high-quality mesh generation method for a rotating shaft magnetic liquid sealing structure, which comprises the following steps: constructing a two-dimensional axisymmetric magnetic liquid sealing model; importing the two-dimensional axisymmetric magnetic liquid sealing model into a grid division tool for grid division to obtain a target grid file; and detecting the precision of each grid in the target grid file, and if a preset precision index is met, outputting the target grid file. According to the method, an encryption area needing high-precision calculation and a relatively secondary non-encryption area are accurately defined based on a regional partitioning strategy, so that unnecessary calculation burden caused by global encryption is effectively avoided; meanwhile, according to the characteristics that the size of the encrypted area is small and the magnetic flux density changes extremely violently, a two-in-one grid transition strategy is innovatively introduced into the junction of the encrypted area and the non-encrypted area, and the structured multi-scale grid layout is achieved while it is ensured that the grids are smoothly transited from high density to low density.

The invention provides a range radar device based on ultra-wideband resolvable chaotic synchronization, which relates to the technical field of radar range finding and comprises a first ultra-wideband resolvable chaotic signal generation circuit unit, a second digital excitation circuit unit, a third ultra-wideband resolvable chaotic signal synchronization circuit unit and a fourth wideband related receiving circuit unit. According to the invention, the problem of radar real-time ranging with high resolution and high measurement signal-to-noise ratio is solved, high-resolution real-time correlation ranging is realized through the generation circuit and the receiving circuit of the ultra-wideband analyzable chaotic signal, correlation receiving is completed only by simulating a matched filter, and high-sampling-rate data pressure and large correlation calculation amount are avoided. In addition, a receiving end can analyze synchronization of chaotic signals through driving of a digital excitation circuit, and chaotic echo signals are accurately reconstructed to improve the signal-to-noise ratio of distance measurement.

The present application relates to a kind of high voltage direct current transmission inverter side receiving end system small signal stability evaluation method, system, electronic equipment and computer readable storage medium, wherein method includes: gathering commutation bus steady-state operating parameter to construct inverter side nonlinear control equation group;Small disturbance response characteristics are extracted at system steady-state operating point, and linear equation group is established;Linear equation group is decoupled in combination with feedback control loop parameter, and aggregated to build transfer function admittance model;Based on the model, time-domain dynamic response curve is obtained, and the small signal stability evaluation index of system is output.The present application avoids complex electromagnetic transient modeling process, can intuitively, accurately reflect the time sequence dynamic characteristics of system under actual power grid fluctuation, greatly improve the efficiency of small signal stability analysis and evaluation of direct current transmission system.

The invention discloses a satellite attitude super-agile sweep neural network minimum parameter learning sliding mode control method, which belongs to the technical field of microsatellite attitude control, and comprises the following steps: S1, in a satellite control process, establishing an actual output model of a neural network; s2, introducing a minimum parameter learning method to replace weight adjustment of a radial basis function neural network, and converting calculation of a neural network weight matrix W into solution of a single parameter # imgabs0 #; according to the method, the system state is converged to zero within finite time by using fast Terminal sliding mode control, and a switching item in a sliding mode is approached through a radial basis function neural network, so that the aim of weakening buffeting is fulfilled; meanwhile, a minimum parameter learning method is designed to replace weight adjustment of the radial basis function neural network, a large amount of operation of a weight matrix is avoided, the efficiency of the controller is effectively improved, and the method has the advantages of being good in convergence, imaging and control effect.

The invention discloses a two-stage compression denoising imaging method for a large-area array Geiger mode APD (avalanche photo diode). The method comprises the following steps: intercepting photon events in a pre-sampling time window at the front part of a data block to carry out global histogram statistics, fitting and estimating time domain characteristics of background noise based on Poisson distribution, screening and compressing full-frame data, and eliminating most noise; carrying out local joint histogram statistics and peak value extraction on the compressed data based on a pixel space neighborhood, and reconstructing a preliminary distance image and a preliminary intensity image; if the effective data in the neighborhood of the data meets the flat area condition, standard median filtering is executed; otherwise, starting a robust filtering algorithm based on absolute median difference MAD for processing. According to the method, the cooperation of front-end data efficient compression and rear-end accurate denoising and edge preserving is realized, and the method has the advantages of high compression ratio, high adaptability and excellent imaging quality, and is particularly suitable for real-time three-dimensional imaging of a large-area-array GM-APD laser radar system.

The present application relates to a processing method, device, computer equipment, computer-readable storage medium and computer program product for a neural network model. The method comprises: obtaining an embedding matrix corresponding to an input sequence through an embedding representation layer of a target neural network model; obtaining the characteristic dimension of the embedding matrix and the characteristic dimension of the weight matrix of the attention layer when obtaining the attention weight corresponding to the embedding matrix through the attention layer of the target neural network model; when the characteristic dimension of the embedding matrix and the characteristic dimension of the weight matrix satisfy a preset relationship, reading a pre-calculated factor matrix from a local cache, and calculating the attention weight corresponding to the embedding matrix according to the embedding matrix and the pre-calculated factor matrix; the weight matrix includes a query matrix and a key matrix, and the pre-calculated factor matrix is ​​a matrix obtained by performing a dot product operation on the query matrix and the key matrix, which can take into account both the amount of calculation and the amount of parameters in the model calculation process.

The application discloses a method for measuring catalyst pore defects, which comprises the following steps: mixing molten paraffin with a catalyst, stirring and ultrasonic dispersing simultaneously; rapidly cooling the mixture of the paraffin and the catalyst by using liquid nitrogen; analyzing the three-dimensional structure of the catalyst by using X-ray computed tomography; processing the three-dimensional scanning original data by using data processing software, and the program is sequentially median filtering, threshold segmentation, boundary particle removal, filling the catalyst image holes by using a morphological hole filling algorithm, particle marking and screening, taking an image difference set, and statistically calculating data, and finally obtaining the catalyst pore rate. Compared with the traditional measurement method, the method can more accurately observe the catalyst structure from the three-dimensional angle, and provides data for the manufacture and evaluation of the catalyst.

The invention discloses a multi-source vehicle speed fusion method based on a hunting algorithm to optimize a radial basis neural network, which includes: obtaining the location vehicle speed of a certain section of a target road section through three location vehicle speed collection methods, and dividing the location vehicle speed into a fusion group and a control group; clustering the fusion group using K-means clustering, and determining the optimal cluster center number K by the elbow method; using the optimal cluster center number K as the number of neurons of the radial basis neural network; constructing the radial basis neural network using a Gaussian function as a kernel function according to the number of neurons; using the fusion group as the input of the radial basis neural network, comparing the output of the neural network with the corresponding control group, and using the comparison result as a fitness function; (S5) using a hunting algorithm, the fitness function optimizes the parameters of the radial basis neural network to obtain the parameters of the radial basis neural network, and using the radial basis neural network to fuse multi-source vehicle speeds. The method of the invention can accurately and efficiently fuse the location vehicle speeds.